# Eve gap analysis: what Eve should borrow from Eunomia and Linear

- Date: 2026-09-22; revised 2026-09-29 on reviews 79d73b2d and 6d51a5df
- Baseline: oshun `10771d1e8e4` (branch `olt67`); eunomia `FEATURES.md` as read,
  unpinned, on 2026-09-22 (506 checked, 31 open, the counts of the file as
  committed at `4c397fac7e53` on 2026-09-22), checked by the author on
  2026-09-29 at eunomia `75f621cf59145167cbfc14055458260cb80a250f` (510 checked, 32 open; section
  14.1, not independently verified: the repository is private); Linear public
  changelog through 2026-09-14, cited by direct page (section 14)
- Author: Claude Fable 5.1, at Elikem's request
- Status: promoted. Every actionable item is a task in Eve board family
  `eve-gap-analysis` (priority P0, rank 1) as of 2026-09-22; section 15 maps
  items to task keys. The report stays the reasoning record; the board is the
  backlog of record.
- Revision of 2026-09-29 (claude-board-worker, review 79d73b2d, decision
  5be600c3): Eunomia claims are checked by the author against a pinned commit,
  with excerpts and file blob ids (section 14.1; not independently verified,
  since the repository is private and outside the review package); every Zel-dependent passage and row is labelled as reported
  and unverifiable in this review, and the recommendations that rest only on
  Zel are marked provisional; Linear claims cite the pages that support them;
  retired repository files are cited by Eve document ID and revision; section
  13.1 now records the product graph's existing Eve task section; the check's
  question count is stated per path; the board totals quote a dated capture of
  `./eve board`, unverified by review.
- Revision of 2026-09-29 on review 6d51a5df: G8's orphan modules are those no
  declared root reaches, not those with no importer, and R9's floor uses that
  definition; section 12's guidance row and R34 describe Linear Agent guidance
  and Triage Intelligence guidance separately, and R34's family-over-global
  rule is marked as Eve's design choice; section 14.1 and the board totals say
  who checked them and that the review could not.
- Revision of 2026-09-29 on review 92e73a5c (needs evidence): the cited
  Eunomia text is supplied as two Eve evidence documents, relations of this
  one: `96edcbf0-d40d-4fe5-a526-19dfd18e3bee` (revision 1: `FEATURES.md` windows, checkbox counts and the 32
  open items) and `63fa87db-c65f-4c4b-93a3-9defa964ed81` (revision 1: the other 78 cited files and the
  `docs/v2/adr/` listing). Each is a verbatim copy, made by the author, of
  every cited line with three lines either side. Section 15 no longer states
  board totals: the author's capture could not be checked by review, and the
  document's argument does not rest on them. The Zel limitation is unchanged
  (decision 5be600c3): no Zel revision is available to supply.
- Revision of 2026-09-29 on review d9a39739 (needs evidence) and decision
  db2f4572: Zel is excluded from this analysis because its source revision is
  unavailable for review. The exclusion does not imply that Zel lacks any
  capability earlier revisions reported. Sections 4 and 11 are now short
  exclusion notes, and every Zel row, bullet and rationale elsewhere is
  removed. Of the items earlier marked provisional on Zel, R13 is retained on
  ADR-0082 and R30 on the SMX decks' record of empty-reply provider tails
  (narrowed to completion recovery on an empty reply); R20, R31 and R32 are
  removed, as are the checks C01 to C03 and the later items L02 and L03, whose
  only source was the Zel sweep (section 15). The Eunomia additions and folds
  that section 11 carried (R25 to R29) now close section 10. The historical
  analysis with the Zel material is revision 14 of this document (Eve document
  `e1809ed1-1365-8943-b0f7-4220910e4a51`) and the Git history of
  `docs/research/gap-analysis-eve-vs-eunomia-zel-linear.md`.
- Same revision, the rest of review d9a39739 and decision db2f4572: the
  Eunomia evidence is now whole where a claim needs it. `FEATURES.md` is
  supplied in full (four parts whose counts reproduce 510 checked and 32
  open), with every tracked path at the commit, the code behind each finding
  (38 numbered items: files, line ranges and `git grep` output), and a
  claim-to-source map (section 14.1). The status "verified" becomes
  "recorded": a claim rests on what `FEATURES.md`, the ADRs, the cited lines
  or a code item show, never on an untraced reading of the code, and no test
  was run. Code findings the evidence does not show are removed: 14.1's claim
  that sources are declared only by four composers was wider than the
  repository (a `sources=` search finds more callers), and section 10's
  artifact-lineage row no longer says PRDs are registered without sources.
  Section 9 no longer cites the polish defect ledger, whose count
  nothing else rested on; its relation is dropped. A read-only resolution
  report of every Eve document this one names or pins is `73fa2c97-2b38-45db-9ca1-70d572db1c5c`.
- Revision of 2026-09-29 on review 2ee9bce4 (three minor findings): the
  comparison tables' labels read "recorded", as 14.1 defines it, and the rows
  whose runtime wiring was not traced say so (ClarificationIssue's gate
  blocking and re-fire, the State of Affairs capture, the lease callers);
  section 12's Loops row gives Linear's two exceptions (scheduled loops are
  outside the write-to-trigger setting; loops that only apply configured
  issue updates run without AI credits), and R5 and 13.2 mark Eve's broader
  rule as Eve's design choice; the agent sessions row gives Linear's actual
  ten-second rule (any activity or an external URL update), and R35 marks its
  thought-first deadline as Eve's choice.

Thirty-three candidate features (R1–R36 less R20, R31 and R32, removed in the
revision of 2026-09-29), eleven UI recommendations (U1–U11)
and thirteen graph capabilities (G1–G13),
ranked for a one-person company: Elikem, plus an open-ended fleet of agents
that Eve runs. Every item names where it comes from, what Eve already has, and
where it would land in the repository. Section 9 assesses Eve's own UI and
dashboards; section 10 records the second, source-level sweep of Eunomia;
section 12 is the second pass on Linear's 2026 features; section 13 sets out
graphs as Eve's spine. Sections 4 and 11 are exclusion notes, kept so that
later section numbers stay stable.

## 1. The frame: Elikem, plus agents

Elikem is the only human at Oshun Synthetics, and Eve exists so that Elikem can
design, plan, build, create content, manage and run the whole company alone.
That single fact reorders every comparison below. Eunomia and Linear are
both built for teams: reviewer roles, tester quorums, workspaces, seats, share
links. None of that is a gap for Eve. What matters instead is three things.

- **Elikem's attention is the scarcest resource in the system.** Every feature
  that keeps low-value decisions away from Elikem, or presents the ones that do
  reach Elikem in a ranked queue, pays back immediately.
- **Agents must be able to act under policy, with audit, when Elikem is not
  there.** Eunomia's delegated-authority Steward is the clearest model of this
  in the two systems.
- **Uncertainty needs one typed channel.** Today an agent that is stuck parks a
  task with a reason. A structured question that blocks the right thing, reaches
  Elikem on the right device, and re-fires the agent when answered is the
  highest-leverage missing piece.

Verdict vocabulary used in the tables:

| Verdict | Meaning |
| --- | --- |
| HAS | Eve already ships this; not worth rebuilding |
| PARTIAL | A related seam exists but the feature as the other system ships it does not |
| GAP | Nothing comparable in Eve |
| SKIP | Exists elsewhere but is wrong for a company of one |

## 2. What Eve already is

Eve is six planes stitched together. Every gap below is measured from this
baseline.

| Plane | Where | What ships today |
| --- | --- | --- |
| Task Board | `tools/eve-task-board`, `infra/hetzner/eve` | Shared PostgreSQL backlog of record. `./eve next / claim / start / done / park / verify / dep / context / prompt / promote / pull`. Leases with fencing tokens and four-hour TTL, journals, families, dependency DAG with cycle refusal, parked reasons, evidence capped at 160 words, `check` (eight questions in `lib/check.mjs`, ten on the shared PostgreSQL board, where `lib/board-check.mjs` adds a stored-policy and a dependency-target check), `drift`, task authoring API with revision checks. HTTPS worker API with scoped keys (`board:read/work/author/review`), interrupted-operation recovery, retry identities. |
| Reviews and automation | `eve_api.reviews`, `tools/eve-reviewer` | Independent review lifecycle (submit, claim, ten-minute lease, complete with approval / changes / missing evidence, reconcile), trusted CI evidence per commit, automatic worker and reviewer switches (Codex `exec` only, default off), presentation proposals. |
| Content and publications | `apps/oshun/eve-content` | The Eve Workspace UI (Tasks, Docs Center, PRDs, Decisions, Architecture, Presentation Center, Review queue, Publications, Media, Automatic agents) over Payload-backed documents with revisions, sections, relations, successors, patch-block edits and review states; sources and task-sources; media with checksum-verified restore; publications with freeze, promote and export; UAT read-only snapshots; 16,379 catalogued documents with a full-text cache and `docs stale`. |
| Workbench | `apps/oshun/bff/src/workbench`, `apps/oshun/bff/src/routes/workbench.ts` | Governed work items (`draft → triaged → ready → leased → in-progress → in-review → shipped → verified`, plus `parked` and `rejected`), queue policy with cloud spend approvals, briefs, reports, triage, verification-failure triage, planning goals / plans / revisions, planning fleet, threads, decisions with ADR export, feature proposals, content briefs and dispatch, release notes (global feed), shipped narratives, `what_shipped_since`, fleet read model, lease listing, graph neighbourhood and explorer, operator labour accounting, publish consents. |
| Assistant runtime | `apps/oshun/bff/src/assistant`, `libs/oshun/assistant` | Builder assistant with exact confirmation cards and reversal, curated tours, drawer voice, contextual invocation, page vision, operator memory (default-on), SMX task-family skills with a prompt-hash ratchet and 251-case evaluation decks, weekly model lifecycle on a price-sorted fp8 OpenRouter route, capacity budget, turn cancellation, history compaction, citation / count / dispatch claim checks, admin tools (incidents, crash groups, review queue, support, moderation, release readiness, model registry, `board_next/show/search`). Library layer: cron and one-shot scheduler, watcher contract and engine (no UI yet, ADR-0081), Discord / Slack / Telegram / Signal / email bridges, subagents with terminal sessions, browser / code-exec / filesystem / vision tools, Voyager skill store, native desktop computer use behind a run policy (ADR-0080). |
| Reliability and evidence | ADR-0080…0088 (Eve documents; files under `docs/adr/` at `10771d1e8e4`, see this document's relations), `tools/eve-everywhere` | Twelve-plane SLO contract, operation-class reliability policy, threat model, red-team matrix, backup and recovery proofs, delivery fault game, roughly 200 evidence and verification scripts. |

Two observations shape the rest of the report. Eve's governance depth
(confirmation proof, fencing, prompt ratchets, SLO contracts) is already ahead
of both comparators. Eve's planning and attention layers are thin: tasks
are still authored by hand from markdown trackers, uncertainty is expressed as a
parked task, and the only ranked view of Elikem's day is the terminal.

## 3. Eunomia: the pipeline and the Steward

Eunomia (`/mnt/volume300/workspace/eunomia` on the Mac; `git@github.com:GreyChimp/eunomia.git`, checked at `75f621cf` in section 14.1) is a Python master/worker system
that takes a project from idea to certified completion: PRDComposer,
TechSpecComposer, EpicPlanner, StoryPlanner, AtomicPlanner, Sequencer,
DeveloperAgent, AgentReviewer, QA, and a certification committee, with four
human gates. It reaches models only through the Claude and Codex CLIs. The
design decisions that transfer to Eve are the structural ones, not the team
ones.

| Eunomia feature | What it does | Eve today | Verdict |
| --- | --- | --- | --- |
| ClarificationIssue (D17) | The only channel for agent uncertainty. Typed scope, concern kind, suggested resolutions, state machine open → answered → applied → resolved (the transitions enforced in `clarification/lifecycle.py`), gate-blocking and scoped re-fire of the emitting agent once the workspace gains an answer (both as FEATURES.md records them, CI-014 and CI-015; their wiring was not traced, 14.1). | `./eve park --reason human` plus a journal note. No state machine, no re-fire, no blocking semantics beyond the parked task. | GAP · recorded, 14.1 |
| Project Steward delegated authority (ADR-0040) | Per-project policy presets, shadow mode, confidence floor, mandatory escalation for high-risk classes, a durable, idempotent decision audit with policy snapshot (its rows carry a run state and retry history, so it is not immutable). The Steward can approve gates, certify tasks, triage bugs, recover failed tasks. | Automatic worker and reviewer switches; publish consents; verify must be "somebody else". No policy object that lets Eve decide on Elikem's behalf inside bounds. | GAP · corrected, 14.1 |
| Composer chain (PRD → spec → epics → stories → atomic tasks) | Each layer emits typed artifacts with acceptance criteria, spec anchors, scope globs, T-shirt sizes, dependency DAG, sibling scope-overlap checks and clarifications for anything thin. | Planning goals, plans and revisions exist in the workbench; board tasks are imported from markdown trackers or authored by hand. | PARTIAL · recorded, 14.1 |
| Cascade barrier (D3/D4) and identity hashes (D16) | A changed PRD or spec marks downstream scopes, pauses in-flight work, re-plans, supersedes, and dispatches. Stable feature identities are specified to survive the re-plan (D16); at `75f621cf` only Layer 1 identity hashing is built, and diffing identities across runs (WS-010) is open. | Document successors and relations; `docs stale` is a 35-second git-blame heuristic run on demand. | PARTIAL · corrected, 14.1 |
| WorkProduct (D15) | Structured output of a developer run: diff set, scope check, test delta, decision log, human inputs, agent notes. The reviewer reads the decision log first. | `done --evidence` takes prose under 160 words and a commit SHA. | GAP · recorded, 14.1 |
| Certification committee (ADR-0043) | Deterministic floor (build, imports, routes, orphans, single DB registry, whole-app render), model ceiling (structural, security), intent tier (feature audit, coherence). Findings dedupe by hybrid identity; loop terminates on novelty with a per-attempt budget; only the floor gates. | `./eve check` asks ten questions of the shared board (eight in `lib/check.mjs`); ~200 evidence scripts under `tools/eve-everywhere` exist but are not a committee that runs per family or release. | PARTIAL · recorded, 14.1 |
| BugReport pipeline and Auto-mapper | Five intake channels, severity classification, embedding rank against features, vision check on evidence, confidence with alternatives, and a human triager who confirms every mapping before a task is derived (low confidence only means the triager starts without a proposed mapping). | Admin crash groups, incidents, support and moderation queues are readable by the assistant; nothing maps them onto board families automatically. | PARTIAL · corrected, 14.1 |
| Progress snapshots and burn-down (ADR-0022) | Append-only snapshots on cron, cascade and gate events; velocity and ETA with confidence tiers; sparklines per project. | `./eve stats` and `./eve board` print current counts only. | GAP · recorded, 14.1 |
| LLM control plane (LLM-CP, CLI-007…010) | Durable invocation ledger, scoped allowances per organisation / project / workload with periods and hard stops, budget-aware routing, fallback chains with account cooldowns, prompt version linked to downstream outcome. | Capacity budget, provider route governance, weekly SMX lifecycle, denial-of-wallet audit. No per-lane allowance periods, no fallback chain across Codex accounts. | PARTIAL · recorded, 14.1 |
| Governed agent configuration (ADR-0045) | Immutable agent revisions, skill version pins, prompt lineage with evaluation, promotion and rollback. | SMX prompt hash ratchet and evaluation decks cover the served bytes; ADR-0082 defines skill ownership. Promotion and rollback are manual re-stamps. | PARTIAL · recorded, 14.1 |
| Master/worker distributed execution (ADR-0019) | Redis Streams dispatch, heartbeats, stale-worker watchdog and job reclaim, bare-clone cache with worktrees, async push after every agent commit. | Worker API, leases, fencing tokens, planning fleet and fleet read model. No heartbeat or automatic reclaim of a dead worker's lease beyond TTL expiry. | PARTIAL · recorded, 14.1 |
| Environment sessions (ADR-0024) | Hosted code-server IDE on EC2, artifact-only evidence workspaces, signed artifact URLs. | None. Review happens on the Mac or the Linux box. | GAP · recorded, 14.1 |
| Notes, brainstorm boards, readiness signal | Draft → promote notes, Excalidraw boards with semantic contracts, agent-generated architecture boards, a green/amber/red readiness hint before drafting a PRD. | Threads and a graph explorer in the workbench; documents in Eve content. | PARTIAL · recorded, 14.1 |
| Work intake (ADR-0038) | A work request is routed deterministically: direct task, additive PRD, or full composition; advisory duplicate detection never suppresses work. | `create_work_item` and `./eve promote`. No routing, no duplicate check. | PARTIAL · recorded, 14.1 |
| Chatbot with mutation gating (D11) | Read tools run inline; any mutation needs confirmation. | Eve's confirmation cards are stronger: complete-state proof before, after parking, and after decline. | HAS · recorded, 14.1 |
| F-classified failures | Every failure carries a code; returning nothing is forbidden. | CLI refusals carry a code and a sentence; the reliability policy classifies operations. | HAS · recorded, 14.1 |
| QA Playbook with two independent testers (ADR-0047) | Blind passes, quorum, tester invitations, per-scenario results. | Not applicable: Elikem is the only tester. | SKIP · recorded, 14.1 |
| Design Studio parity (ADR-0046) | Claude-Design-class studio with design-system library, share links, PDF export. | Oshun has its own studio surfaces; not an Eve concern. | SKIP · recorded, 14.1 |
| Identity, organisations, reviewer roles (D20) | Users, orgs, participant roles, magic links. | Scoped API keys per agent identity are the right shape when Elikem is the only person. | SKIP · recorded, 14.1 |

## 4. Zel: excluded

Zel's source revision is unavailable for review: on 2026-09-29 its repository
could not be retrieved under any of the names tried, and no checkout exists on
the host that made this revision (section 14). This analysis therefore makes
no claim about Zel's features, and nothing here implies that Zel lacks any
capability. The historical 2026-09-22 read of Zel is in revision 14 of this
document and in its Git history (see the header).

## 5. Linear: attention design and agent-native tracking

Linear's 2026 direction is the more relevant of the two: it has turned an
issue tracker into a place where agents are first-class workers. Linear Agent
(March), MCP for the agent (April), Code Intelligence (May), Coding Sessions
(June), Loops (July), coding environments and browser testing (August), Priority
Inbox and Loops for product management (September). The team features are
irrelevant here; the attention and automation features are exactly what Elikem
needs.

| Linear feature | What it does | Eve today | Verdict |
| --- | --- | --- | --- |
| Priority Inbox (2026-09-03) | Separates critical notifications from routine ones, customisable by source or filter. | Notifications arrive as terminal output, the drawer, and channel bridges. No single ranked queue. | GAP |
| Triage inbox, triage rules, triage intelligence | Issues from integrations land in triage; rules set team, status, assignee, labels, project, priority in order; the model suggests properties and flags duplicates. | Work items have a `triaged` status and a triage route; there are no rules and no suggestions. | PARTIAL |
| Loops (2026-07-20, extended 2026-09-14) | Plain-language recurring agent workflows on a schedule or an event; update projects, edit documents, post to Slack. | Watcher contract and scheduler in the library; no product. | PARTIAL |
| Initiatives, project health, project updates, milestones, graphs | Status, priority, target date, health rolled up from the latest update; "Write with Agent" drafts the update from recent changes. | Families are ranked; `what_shipped_since` and shipped narratives exist; no health, target dates or drafted updates. | PARTIAL |
| Cycles | Fixed cadence, cooldowns, auto-add of started work, rollover, capacity dial from the last three cycles, scope-change graph. | None. | GAP |
| Coding Sessions, environments, browser testing, guided reviews | Assign an issue to the agent; a cloud sandbox produces a draft PR with preview links and before / after screenshots; guided reviews organise a diff into explained sections; spend limits per user and timeframe; signed commits. | Automatic worker lane with Codex, cloud spend approvals on leases, publish consents. No preview link, no guided review, no browser-tested evidence attached to the item. | PARTIAL |
| Linear Agent in comments, Slack, and as skills | `@Linear` in any thread; saved skills as slash commands; drafting issues and projects from conversation; agent-assisted document editing with text attribution and checkpoints. | Threads exist; the assistant is invoked from the drawer or contextually. No mention-invocation in threads or documents, no attribution of agent-written spans. | PARTIAL |
| GitHub automations | Branch names and PR links move issues through states automatically. | `done --commit` records a SHA after the fact. | GAP |
| SLAs | Rules apply deadlines by priority; risk indicators and notifications before breach. | Lease TTLs on tasks and reviews; nothing ages a question waiting on Elikem. | PARTIAL |
| Views, display options, insights | Saved filtered views as list, board or timeline; instant breakdowns on any view. | The Eve Workspace Tasks view is a filtered list with a detail pane; the work-item kanban is in the member-shell explorer; `@oshun/workbench-kit` ships DataTable, StageBoard, TimelineEditor and a command palette that the Workspace does not mount. No saved views, no timeline, no breakdowns. | PARTIAL |
| Customer requests | Requests captured from support channels link to issues; customers carry tier and revenue; views by request count. | Support queue, feedback store, member complaints. Release feed is deliberately global. | PARTIAL |
| Asks | Requests from Slack, email and web forms become issues (Linear Agent separately creates issues from Microsoft Teams). | Channel bridges exist; no intake into the board. | PARTIAL |
| Issue relations, sub-issues, templates, callouts | Blocks / blocked-by / related / duplicate with merge; templates for issues, projects and documents. | Dependencies and sections; no duplicate or related relation, no templates. | PARTIAL |
| Release pipeline changelogs | Aggregated release notes with auto-generation. | Release notes and shipped narratives. | HAS |
| Document version history and checkpoints | Restore by checkpoint. | Document revisions with preconditions. | HAS |
| Teams, private sub-teams, team initiatives, third-party app approvals, Okta-managed MCP, seats | Organisation features. | Not applicable. | SKIP |

The pages each row rests on (fetched on 2026-09-29; section 14):

- Priority Inbox: <https://linear.app/changelog/2026-09-03-priority-inbox>, <https://linear.app/docs/inbox>
- Triage inbox, rules and intelligence: <https://linear.app/docs/triage>, <https://linear.app/docs/triage-intelligence>
- Loops: <https://linear.app/changelog/2026-07-20-introducing-loops>, <https://linear.app/changelog/2026-09-14-loops-for-product-management>, <https://linear.app/docs/loops>
- Initiatives, project health, updates, milestones and graphs: <https://linear.app/docs/initiatives>, <https://linear.app/changelog/2026-06-18-agent-assisted-project-updates>, <https://linear.app/docs/project-graph>
- Cycles: <https://linear.app/docs/use-cycles>
- Coding Sessions, environments, browser testing, guided reviews: <https://linear.app/changelog/2026-06-11-coding-sessions>, <https://linear.app/changelog/2026-08-20-coding-environments>, <https://linear.app/docs/coding-sessions>, <https://linear.app/docs/diffs>
- Linear Agent in comments, Slack and as skills: <https://linear.app/changelog/2026-03-24-introducing-linear-agent>, <https://linear.app/docs/linear-agent>, <https://linear.app/changelog/2026-06-04-team-documents>, <https://linear.app/changelog/2026-07-23-agent-assisted-editing>
- GitHub automations: <https://linear.app/docs/github-integration>
- SLAs: <https://linear.app/docs/sla>
- Views, display options, insights: <https://linear.app/docs/custom-views>, <https://linear.app/docs/display-options>, <https://linear.app/docs/insights>
- Customer requests: <https://linear.app/docs/customer-requests>
- Asks: <https://linear.app/docs/linear-asks>, <https://linear.app/changelog/2026-04-02-web-forms-for-linear-asks>
- Issue relations, sub-issues, templates: <https://linear.app/docs/issue-relations>
- Release pipeline changelogs: <https://linear.app/changelog/2026-04-30-releases>, <https://linear.app/docs/releases>
- Document version history and checkpoints: <https://linear.app/changelog/2026-07-23-agent-assisted-editing>

## 6. Recommended build list

Three tiers. NOW items reduce load on Elikem or let Eve act safely while Elikem
is elsewhere. NEXT items deepen planning and delivery quality. LATER items are
real but can wait for the first two tiers to prove out. Effort is relative to
the existing seams: S is days, M is a week or two, L is a multi-week
initiative. Sizes are judgement, not measurements.

### NOW: attention and delegated action

**R1 · Clarification queue · M.** One typed channel for agent uncertainty. An
agent emits a question with scope (task, document section, family), concern
kind, suggested answers and what it blocks. States open → answered → applied →
resolved. Answering re-fires the agent on the same task with the answer in its
brief. Delivered to the drawer and to Telegram or Signal, with an age indicator.
Replaces `park --reason human` for questions.
From: Eunomia D17, Linear SLAs.
Where: `tools/eve-task-board` schema and worker API (`./eve ask`,
`./eve answer`), an MCP tool, `libs/oshun/assistant/src/channels`.

**R2 · Priority inbox for Elikem · M.** A single ranked queue that merges
clarifications, reviews awaiting Elikem, confirmation cards, publish consents,
watcher deliveries, crash groups and lease anomalies. Rules by source and
severity decide what is critical; everything else is a daily digest. Mobile
through the admin-mobile notifications surface and the channel bridges.
From: Linear Priority Inbox, Eunomia two-pane Inbox.
Where: `apps/oshun/web/src/components/workbench/NotificationCentre.tsx`, a BFF
inbox read model, `apps/oshun/admin-mobile`.

**R3 · Eve Steward: delegated authority under policy · M–L.** A versioned
per-family policy with presets (advisory, shadow, delegated), a confidence
floor, a list of decision classes Eve may take alone (verify a shipped task,
approve a non-PRD document review, publish a release note, unpark, retry a
failed lane) and classes that always escalate to R1. Every decision writes an
immutable audit row with the policy snapshot, model, rationale and confidence
(Eunomia's audit is durable and idempotent, not immutable; immutability is this
item's choice).
Shadow mode records what Eve would have decided so Elikem can calibrate before
delegating.
From: Eunomia ADR-0040 and PST-001…010.
Where: `apps/oshun/bff/src/workbench` beside intent machines and publish
consents; automation switches in `eve_api.agent_automation`.

**R4 · Triage before Elikem · M.** New items from crash groups, support,
watchers and channel intake enter as `draft`. Ordered deterministic rules set
family, section and priority. An embedding pass against the task index proposes
the family and flags likely duplicates with confidence and alternatives. Only
what the rules could not settle reaches Elikem through R2.
From: Linear triage rules and intelligence, Eunomia BR Auto-mapper.
Where: `POST /v1/workbench/items/:id/triage`, the board `triaged` state,
`.cache/eve-task-board/knowledge.sqlite` for the embedding index.

**R5 · Eve Loops · L.** Plain-language recurring automations with a cron or
event trigger, bounded tools, wait and approve steps, and a delivery target.
First loops: the weekly family update, the stale-document sweep, lease-expiry
reclaim, the SMX model lifecycle run, a nightly board `check`. Needs the worker
ADR-0081 deferred, plus a list page with pause and run-now.
From: Linear Loops, Eve's own watcher contract.
Where: `libs/oshun/assistant/src/scheduling` and `watchers`, a BFF loop
registry, a `WorkbenchShell` page.

### NEXT: planning and delivery depth

**R6 · Planning chain from an Eve document · L.** From a PRD or design document
in Eve content, generate epics, stories and board tasks with scope globs,
acceptance criteria, spec anchors as Eve section IDs pinned to a revision, a
dependency DAG, sibling scope-overlap checks and a one-day granularity rule.
Thin or contradictory nodes raise R1 questions. Elikem's single approval of a
plan revision dispatches it.
From: Eunomia PRD / TSC / EPL / STL / ATL and Sequencer.
Where: planning goals / plans / revisions in
`apps/oshun/bff/src/routes/planning-*.ts`, the task authoring API,
`./eve promote`.

**R7 · Change cascade from document revisions · M–L.** A superseding revision
signals every task that cites the changed section, pauses leased work on it,
re-plans the scope, supersedes what no longer applies and dispatches the rest.
Stable task identities survive. Make `docs stale` event-driven instead of a
git-blame sweep.
From: Eunomia D3, D4, D16. At `75f621cf` the cascade is built and D16 is Layer 1
hashing only; stable identity across a re-plan is this item's requirement, not
built Eunomia behaviour (section 14.1).
Where: `doc_links`, `task_links`, the content-events collection, intent
machines.

**R8 · Typed work product on done · M.** `./eve done` accepts a structured
product beside the prose: diff set, scope check against the paths the task
declared, test delta, decision log, and the inputs Elikem gave it along the
way. The reviewer reads the decision log first. Eve renders a guided review for
Elikem: sections of the diff with a one-line explanation each.
From: Eunomia D15, Linear Guided Reviews.
Where: `lib/done.mjs`, the worker API, `tools/eve-reviewer`, the review report
schema.

**R9 · Certification committee per family and release · L.** Unify the evidence
scripts into a committee: a deterministic floor (build, typecheck, import
graph, route registration, orphan modules as G8 defines them, whole-app
render), a model ceiling
(structural and security sweeps) and an intent tier (does the family do what
its tracker asked). Findings dedupe by identity and become board tasks; the
loop ends on novelty with a per-attempt budget; only the floor gates a family's
closure.
From: Eunomia ADR-0043, AUD-001…040.
Where: `tools/eve-everywhere/verify-*`, `./eve check`, family policies.

**R10 · Family health and agent-drafted updates · M.** Families gain status,
health (on track, at risk, off track), target dates and milestones. Append-only
progress snapshots on cron and on state changes; velocity and ETA with
confidence tiers; sparklines in the board view. Every week Eve drafts the
family update from `what_shipped_since`, journals and open R1 questions, and
posts it to the release feed and Elikem's channel.
From: Linear initiatives and project updates, Eunomia ADR-0022.
Where: the `families` table, a snapshots table, shipped narratives, release
notes.

**R11 · Commit and PR automation · S–M.** A commit trailer or branch prefix
carries the board key. A push moves the task to in-progress, a merge to main
records the SHA and moves it to shipped, and a webhook refuses `done` for a
commit that does not exist on the remote. Closes the "done before the commit"
defect the board README documents.
From: Linear GitHub automations.
Where: a GitHub webhook into the worker API, `./eve done --commit` validation.

**R12 · Board page: the command centre · M–L.** A browser view of families and
tasks as list, board and timeline with saved filters, keyboard-first
navigation, quick-add into `draft`, and the R2 inbox docked. The Eve Workspace
Tasks view is the starting point (list, filters, detail, actions); the
workbench kit already has DataTable, StageBoard, TimelineEditor, HierarchyTree
and a command palette. Section 9 (U1, U2, U3) details the UI side.
From: Linear views, Eunomia Boards / MyQueue / State of Affairs.
Where: `apps/oshun/web/src/components/workbench`, BFF read-only board tools.

**R30 · Completion recovery on an empty reply · NEXT · S.** Run a bounded
second attempt when an Eve turn ends on an empty reply.
From: Eve's own evidence: the SMX evaluation decks (section 2) record
empty-reply provider tails, and Eve makes no second attempt on an empty reply.
The board epic's working-context recorder, its other retry triggers and its
approval checkpoint are outside this analysis (section 15).
Where: `apps/oshun/bff/src/assistant`, beside `tool-result-digest.ts` and
`history-compaction.ts`.

### LATER: compounding advantages

**R13 · Learned skills from journals · L.** Mine completed task journals, review
reports and Elikem's corrections into candidate recipes; a review queue admits
them under the ADR-0082 contract; activation feedback feeds the next mining
pass. The SMX ratchet stays untouched.
From: Eve's own record: ADR-0082 (an Eve document; section 2) defines an
admission contract, and nothing mines the journals.
Where: `libs/oshun/assistant/src/skills`, the `events` journal, eval decks.

**R14 · Per-lane allowances and fallback chains · M.** Daily, weekly and
monthly allowances per lane with hard stops and alerts; fallback chains with
cooldowns across the Codex subscription and OpenRouter routes; an invocation
ledger that links the served prompt version to the downstream verify outcome.
From: Eunomia CLI-007…010, LLM-CP.
Where: capacity budget, provider route governance, model lifecycle persistence.

**R15 · Governed connectors: Eve as MCP client · M–L.** A connector registry
with encrypted credentials, tool-catalog cache, risk classification and the
existing read-inline / write-as-card rule. First connectors: GitHub, Hetzner,
OpenRouter usage, Stripe, Telegram.
From: Linear Agent MCP.
Where: MCP trust and conformance evidence in `tools/eve-everywhere`, toolset
scope.

**R16 · Cycles for the fleet · M.** A weekly cycle that auto-adds started tasks,
rolls unfinished ones forward and shows a capacity dial from the last three
cycles' measured throughput per lane, with a scope-change graph. Gives Elikem a
pace to plan against instead of an open backlog.
From: Linear cycles.
Where: board schema, R10 snapshots.

**R17 · Work-request intake with routing · M.** A message from Telegram, Signal
or email becomes a work request. Deterministic routing sends it as a direct
task, an addition to an existing family plan, or a new R6 planning run;
duplicate detection is advisory and never suppresses.
From: Eunomia ADR-0038, Linear Asks.
Where: channel bridges, `POST /v1/workbench/items`, R4 triage.

**R18 · Preview links and browser-tested evidence · L.** Each agent PR gets an
ephemeral preview on the Hetzner Dev host and before / after screenshots
attached to the work item, with spend limits per lane per timeframe. Guided
review from R8 links to it.
From: Linear Coding Sessions and environments.
Where: Hetzner release controller, publish consents, cloud spend approvals.

**R19 · Hosted review IDE · M.** code-server on the Linux box behind Eve auth so
agent work can be reviewed from any device, with artifact-only evidence
workspaces for runs that should not expose a full checkout.
From: Eunomia ESP-015…019.
Where: `infra/hetzner/eve`.

**R21 · Notes and brainstorm intake with a readiness signal · M.** A notes kind
and Excalidraw boards in Eve content as planning inputs, promoted into a PRD
run; a green/amber/red readiness hint before R6 drafts anything.
From: Eunomia NOT, BRD, RDY.
Where: Eve content collections, threads.

**R22 · Full-duplex voice · L.** Only if the drawer voice proves limiting for
hands-free operation; the existing voice-depth decision record should be the
gate.
From: Eve's own gate, the voice-depth decision record.
Where: `apps/oshun/bff/src/assistant/voice-config.ts`.

**R23 · Member request counts on tasks · S–M.** Aggregate counts of member
requests per task and family from the support and feedback stores, so Elikem
can sort by demand. Keeps the standing decision that the release feed carries
no per-member targeting.
From: Linear customer requests.
Where: feedback store, support queue, `task_links`.

**R24 · Templates and relations · S.** Task and document templates; duplicate
and related relations with a merge that carries journals and links to the
canonical task.
From: Linear issue relations and templates.
Where: `task_deps`, the task authoring API.

### Suggested order

R1 first, because R3, R4, R5 and R6 all need somewhere to send a question. R2
second, because once questions and reviews have a home Elikem will want them
ranked. R3 third in shadow mode, so that by the time R5 and R6 generate
decisions Elikem already knows how well Eve decides. R11 is the cheapest item on
the list and closes a known defect, so it can slot in anywhere. R29 (pre-publish
Git integration, section 10) is also NOW: it removes the merge work every
session currently does by hand. R34 (standing guidance, section 12) is NOW
because it is small and every later item reads it.

## 7. Deliberately not building

These exist in the comparators and would be wasted effort for a company of one.
Listing them keeps future sessions from re-proposing them.

- Workspaces, organisations, tenants, teams, private sub-teams, invitations,
  seats, per-seat billing.
- Human RBAC UX. Scoped agent credentials are the right model; Elikem holds the
  owner role and there is no second human to authorise.
- Two-tester QA quorums, blind passes, tester invitations, triage responsibility
  rotations, on-call integrations.
- Share links with viewer / commenter / editor roles, third-party app
  approvals, identity-provider-managed MCP.
- Design Studio parity for Eve itself. Oshun's product studios own design work.
- Multi-human text attribution. Attribution of agent-written spans versus
  Elikem's own edits is still useful and folds into R8's review and Eve
  content's revisions.

## 8. Rules worth revisiting

- **Name the human.** Across Eve's prompts, tool descriptions, UI copy, ADRs
  and task text, "the human", "the operator", "the owner" and "a human
  reviewer" all mean one person. Say Elikem. A census of those phrases in
  `apps/oshun/bff/src/assistant`, `libs/oshun/assistant`, `AGENTS.md` and the
  Eve ADRs is a small task and removes a standing ambiguity for every agent that
  reads them.
- **"PRD, ADR and architecture approvals require an independent human"** in
  `AGENTS.md`. With Elikem as the only human, "independent" can only mean "not
  the authoring agent". Suggested wording: Elikem approves; the authoring agent
  never self-approves; a reviewer identity may pre-review. R3 then decides which
  document kinds the Steward may approve alone.
- **Verify must be "somebody else".** Today that is a second agent identity. R3
  should make that explicit: verify is delegated to Eve under policy, with
  Elikem as the escalation path, so the rule stops depending on which credential
  happened to be loaded.
- **The global release feed with no per-member targeting** was a deliberate
  decision on 2026-08-22. R23 respects it by counting rather than targeting.
- **Parking for "human" reasons** should become an R1 question addressed to
  Elikem. The park reason stays for things no agent can obtain: licences,
  devices, accounts, a human act.

## 9. Eve's UI and dashboards: comprehensiveness and ergonomics

Assessed on 2026-09-22 from source at `10771d1e8e4` and from the twelve
showcase frames in `evidence/eve-builder-showcase` (captured 2026-08-29). No
server was booted for this pass. Two prior audits already cover parts of this
ground and are not repeated: the chat-surface parity audit
(chat-surface parity audit (Eve document `3e953362-dfd2-8220-ac55-d8cc967458ef`, revision 1; the file `docs/audits/EVE_CHAT_UI_PARITY_AUDIT_2026-09-11.md` at `10771d1e8e4`), whose 205 closure items
stood at 111 open on 2026-09-16) and
the terminology table (assistant terminology table (Eve document `0ac65a03-7e1f-8606-b0bb-4e83d6990904`, revision 1; the file `docs/audits/EVE_ASSISTANT_TERMINOLOGY_2026-08.md` at `10771d1e8e4`)).

### 9.1 The surfaces Elikem has today

| Surface | App and port | What it holds | Designed for |
| --- | --- | --- | --- |
| Eve Workspace | `apps/oshun/eve-content` (Payload + Next) at `/workspace` | Nav: Tasks, Docs Center, PRDs, Decisions, Architecture, Presentation Center, Review queue, Publications; Media library; Automatic agents switches; the generic Payload admin at `/admin` for the `documents`, `users`, `content-events` and `content-publications` collections. Tasks: search, project (family) and status filters, "mine" toggle, list and detail with claim (30 min), start, extend, release, park, unpark, note, finish work with evidence, verify, add and remove dependencies, "enables", linked documents, an activity journal. Documents: library with search, working documents by kind and state, pinned revisions, read and edit modes, section preview, a review and history inspector, relationships, a presentation editor, diagrams. | A team of authors, workers and reviewers with scoped credentials. Copy says "Your team's work, ownership and evidence." Light theme only, purple accent, responsive to 700 px. |
| Oshun Operations | `apps/oshun/admin` at `:3020` | Seventeen workspaces in four groups (governance, safety, content, operations): dashboard cards, unified inbox (workspace, priority, SLA and assignee filters, bulk assign to an operator id), review, policy, trust and safety, incidents, editorial, personas, models, support, privacy, analytics, Isis operations. The Operations Copilot drawer (about 460 px): context band (workspace, operator, memory), transcript with a `Consulted:` tool trace and read-aloud, confirmation cards with Approve and Decline, dictation, stop, Tours, Session details. | Many operators on shifts. Placeholder text: "Ask about queues, incidents, SLAs…". Operator identity reads "Studio Operator 01". Dark theme only. |
| Workbench explorer | `apps/oshun/web` at `:3010`, route `/assistant/graph`, inside the Lilith customer shell | Product graph with explore, work, decisions and changes tabs; work-item kanban lanes (draft, triaged, ready, leased, in-progress); ADR timeline; coverage line ("19/19 domains verified"). Built on `@oshun/workbench-kit`: three-region shell, command palette, inspector, notification centre, status strip. | Builder use, but mounted in the member product's shell with the member's serif light theme and the Lilith persona. |
| `./eve` and `TODOS/BOARD.md` | terminal | `next`, `show`, `context`, `prompt`, `families`, `stats`, `docs search`, `docs related`, `docs stale`, `check`, `drift`, reviews, promote and pull. | Coding agents first, Elikem second. From this box `./eve board` timed out on 2026-09-22 without an SSH tunnel or an API key file. |
| Oshun Admin mobile | `apps/oshun/admin-mobile` | Urgent, Inbox, Incidents, More: queue health, SLA breaches, incident acknowledgement, review cards. | On-call operators. No board task, document, review or confirmation-card surface. |
| Presentation Center and Docs Center | `docs/presentations/presentation-center` | The Eve learning path (twelve chapters, 6 h 23 m) and Use Eve (six chapters). | Teaching Eve; the accepted "Meet Eve" specimen is the target look for the redesign ledger. |

### 9.2 Comprehensiveness: what the surfaces cover and what they do not

What is covered well:

- Governed writes. Every mutation from the copilot parks an exact confirmation
  card, and the showcase frames prove approve, decline and read-back against
  Postgres. This is the strongest UI behaviour in the system.
- Document lifecycle. Revisions, review findings, publications with freeze and
  promote, UAT read-only mode, and a proper history inspector.
- Task mechanics. Claim, lease, evidence, verify, dependencies and the journal
  are all reachable from the Tasks view with the same fencing rules as the CLI.
- Operations queues. Review, moderation, incidents, support, privacy and
  models each have a workspace and the inbox composes them.

What is missing for Elikem:

1. **There is no home.** Nothing shows the state of the company on one screen:
   open tasks by family, reviews waiting on Elikem, publications in flight,
   agents holding leases, spend this week, incidents. The admin dashboard is a
   grid of operations workspaces; the Eve Workspace opens on the document
   library; the Tasks view shows one totals line.
2. **Families exist only in the terminal.** `./eve families` ranks them; the
   Tasks view has a "project" filter and nothing else. No family page, no
   health, no next-ready list, no counts by state.
3. **The agent fleet is invisible.** `admin_agent_fleet`, `list_agent_leases`,
   the planning fleet routes and the operator labour accounting exist as tools
   and endpoints. The only fleet UI is the two automation switches. Elikem
   cannot see who holds which lease, what each lane spent, or the last failed
   run without asking the copilot.
4. **Two task-shaped things live in two apps.** Board tasks are in the Eve
   Workspace; workbench work items are in the explorer inside the member shell.
   The vocabulary was unified on 2026-09-18 (ETB decision 4); the UI was not.
   The kanban is on the work-item side, the journal and evidence on the board
   side, and `./eve promote` is the only bridge.
5. **Task detail is behind the CLI.** `./eve next` prints `path:line`;
   `./eve context` bundles the documents; `./eve prompt` produces the brief;
   `./eve docs related` lists what a task cites. The Tasks view shows none of
   these, renders the journal as raw JSON in a `<pre>`, and links documents by
   path only.
6. **No attention queue.** The admin inbox is an operations queue with assignee
   and SLA semantics written for a team. Nothing gathers the things only Elikem
   can do: answer a question, approve a card, verify a shipped task, decide a
   publish consent.
7. **The copilot is a side drawer with no history.** No conversation list,
   search, rename or resume; no attachments; an operations placeholder; memory
   defaults to "Session only" in the drawer's context band. The parity audit
   already binds these to tasks; the point here is that the drawer is Eve's
   only conversational surface for building and planning, and it is sized and
   worded for queue questions.
8. **Nothing on mobile is Eve.** Admin mobile handles incidents and reviews. A
   confirmation card, a clarification, a document review or a task journal
   cannot be reached from a phone.

### 9.3 Ergonomics: how the surfaces feel to use

| Surface | Strengths | Weaknesses |
| --- | --- | --- |
| Eve Workspace | Plain, fast, readable; every action names its refusal; revision pinning and the review inspector are exactly right; search on both tasks and documents. | Eight top-level nav links with no hierarchy or counts; no keyboard model (three files handle keys at all); no command palette even though `@oshun/workbench-kit` ships one; light theme only; team copy; the Payload admin is a second, unstyled UI over the same collections; journal entries are unformatted. |
| Oshun Operations | Clear grouping, SLA pills, `Consulted:` traces, tours, read-aloud, ⌘K search and ⌘/ copilot. | Built for shifts: assignee filters, bulk assign to an operator id, "Studio Operator 01"; dark-only; the copilot drawer covers a third of the page and cannot be popped out; three names for one thing on one screen ("Operations Copilot", "Admin assistant", "Copilot"). |
| Workbench explorer | The graph plus work plus decisions plus changes on one canvas is the right idea; the shell's arrangement logic is proof-backed and accessible. | Lives in the customer shell under the Lilith brand with the member's fonts; a builder tool wearing a member costume. Cards carry no board key, evidence or journal. |
| `./eve` | Complete, byte-stable output, refusals with codes, `context` and `prompt` are the best briefs in the system. | Needs a tunnel or key file; `board` output is a text dump; no way to answer a question or approve a card from the terminal because those do not exist yet. |
| Admin mobile | Thumb-reachable tabs, deep links, offline banner. | No Eve function at all. |

Cross-cutting ergonomic problems:

- **Three sign-ins, three shells, three visual languages.** Payload users for
  the Eve Workspace, the admin session for Operations, the member session for
  the explorer, and an API key for the CLI. Purple-on-white, slate-dark and
  cream-serif for one person's tool.
- **Vocabulary drift on operator surfaces.** The 2026-08 terminology table
  fixed member surfaces around Lilith and reserved "copilot" for operator
  surfaces per the brand guide (Eve document `5b3e02e1-e65c-83d8-969d-40e6247463ba`, revision 1; the file `V1/BRAND.md` at `10771d1e8e4`). Operator surfaces now say copilot, admin
  assistant and Eve interchangeably, and never say Elikem.
- **The best information is where Elikem is not.** The briefs, context bundles
  and family ranking are in the CLI; the kanban is in the member shell; the
  queues are in Operations; the documents are in the Workspace.

### 9.4 UI recommendations

These are additions to the build list in section 6 and use the same tiers and
effort scale. U1 and U2 are where R2 and R12 should land.

**U1 · Eve Home · NOW · M.** A single arrival page in the Eve Workspace:
Elikem's attention queue (R2) at the top, then families with health and next
ready task (R10), agents holding leases and this week's spend per lane, reviews
and publish consents waiting, what shipped since Elikem last looked, incidents.
Every tile links into the existing view. Where: `apps/oshun/eve-content/src/workspace`,
reading the board API, the workbench fleet read model and the admin overview.

**U2 · One board · NOW · M–L.** Bring the work-item lanes out of the member
shell and into the Eve Workspace as a board with list, kanban and timeline
views over board tasks and their promoted work items together, with saved
filters and family grouping. The explorer keeps the graph; the lanes move.
Where: mount `@oshun/workbench-kit` primitives (DataTable, StageBoard,
TimelineEditor, command palette) in the Workspace.

**U3 · Task detail parity with the CLI · NOW · S–M.** Show `path:line`, the
`context` bundle, the `prompt` brief with a copy button, related documents,
the current fencing token and lease expiry, the implementation commit as a
link, and a formatted journal instead of JSON. Attach R8's structured evidence
here when it exists.

**U4 · Family pages · NEXT · M.** One page per family: status, health, target
date, open and parked counts, next ready, blocked-by graph, recent journal,
the family's policy text. This is where R3's Steward policy is edited.

**U5 · Fleet page · NEXT · M.** Leases with holder, age and token, automation
switches, per-lane allowance and spend (R14), last runs with outcome, a stop
control per lane. Where: `planning-fleet.ts`, `list_agent_leases`,
`operator-labor-accounting.ts`.

**U6 · Eve as one conversation everywhere · NEXT · M.** A full-page Eve chat
in the Workspace with history, search, rename and resume, plus the same
drawer in Operations and the explorer. One name on every operator surface.
The brand guide (Eve document `5b3e02e1-e65c-83d8-969d-40e6247463ba`, revision 1; the file `V1/BRAND.md` at `10771d1e8e4`) currently reserves "copilot" for operator surfaces; that is
Elikem's call to change, and the report recommends changing it to Eve.

**U7 · One sign-in, one shell · NEXT · M–L.** Either federate the three
sessions or fold the Eve Workspace and the builder explorer into Operations
under one navigation with a "Build" group beside governance, safety, content
and operations. The admin information architecture already has the grouping
model; it lacks the build group.

**U8 · One visual language, both themes · LATER · M.** Adopt the workbench-kit
tokens across the Workspace and Operations, with light and dark in both. The
Presentation Center's "Meet Eve" specimen is the accepted look.

**U9 · Keyboard-first Workspace · LATER · S–M.** Mount the command palette,
add list navigation and single-key actions for claim, note, finish and verify,
and make every card's Approve and Decline reachable by keyboard.

**U10 · Eve on the phone · LATER · M.** Add an Eve tab to admin mobile:
attention queue, confirmation cards, clarification answers, task journal and
document review, reusing the deep-link hydration that exists.

**U11 · Copy census · NOW · S.** Replace team language on Eve surfaces ("Your
team's work", "assigned to me", "Target operator id for bulk assign", "Studio
Operator 01") with Elikem's name and single-person phrasing; one lock in the
string inventory keeps it that way.

## 10. Second pass on Eunomia: source-level sweep

A second sweep walked the `eunomia/` package, its API routes, chatbot tool
registry, YAML registries, the web pages and components, the mobile app and
every ADR through 0047, looking only for things section 3 did not cover. The
findings below are the ones that transfer to a one-person company. Team-only
findings (QA workforce and reviewer tiers, plant seeding, payments, collab
channels, tenant admin, Penpot SSO, Figma bridge) are omitted. "Eve today" is
what this checkout shows; where nothing comparable was found it says so.

| Eunomia feature | What it does | Eve today | Disposition |
| --- | --- | --- | --- |
| Operations projection (`eunomia/operations.py`, `operation_reporter.py`, `api/routes/operations.py`, `docs/v2/operations-inventory.md`) | One durable row per long operation, created before dispatch; real worker phases and heartbeats ("never synthesize percentages without a real total"); sanitised terminal reasons; acknowledge, retry, cancel; SSE stream; watchdog fails abandoned rows. | Turn traces and a telemetry sink exist; workbench-kit ships a `JobQueueView` primitive; there is no cross-job operations table spanning renders, narration, builds and agent runs. | New item R25 · recorded, 14.1 |
| Contextual guide tools (`chatbot/registry.py`: `get_current_screen`, `search_eunomia_help`, `get_recommended_next_steps`, `find_destination`; `web/src/components/guide`) | The assistant knows the current screen's actions, searches a product catalogue, returns sourced next steps and resolves "where do I do X" to a route. | Tours, contextual invocation and page vision exist; there is no per-screen action catalogue or next-step tool. | Fold into U6 · recorded, 14.1 |
| "What is Eunomia doing / why is it stuck" (`get_active_operations`, `get_operation_details`, `explain_blocker`) | Conversational status from persisted operation state. | Nothing comparable. | With R25 · recorded, 14.1 |
| Assistant knowledge-gap records (`api/routes/assistant_feedback.py`, `AssistantInsightsSettingsPage.tsx`) | Thumbs-down and unresolved answers become durable records typed `no_knowledge`, `no_tool`, `permission`, `ambiguous`, `runtime_failure`, `unhelpful`, grouped by screen. | `feedback-store.ts` records feedback; nothing turns it into a typed gap queue. | New item R26 · recorded, 14.1 |
| State of Affairs weekly report (`eunomia/state_of_affairs/`, `StateOfAffairsPage.tsx`) | Per project: progress delta, ETA with confidence, gates pending, open bugs, risk flags, one-line insight; rollups for biggest movers, stalled, watchlist; HTML export; a weekly report period and an on-demand snapshot capture (whether anything schedules a weekly capture was not traced, 14.1). | `what_shipped_since` and shipped narratives. | Fold into R10 · recorded, 14.1 |
| Dashboard "Suggested next move", "How to avoid stalls", "Due soon" (`DashboardPage.tsx`) | Fixed guidance, not a computed pick: one of two set sentences for the next move (chosen by whether an active project exists), set stall-coaching copy, and the projects due in the next 14 days. | Nothing comparable. | Fold into U1; the fold rests on U1's own need for a computed next move, which this panel does not compute · corrected, 14.1 |
| Continue-pipeline verb with a refusal matrix (`eunomia/lifecycle/continue_pipeline.py`) | "Advance this project" dispatches the one unstarted step, or refuses with a named reason (open clarifications, pending gate, failed tasks, no remote, no worker). | Nothing comparable; `./eve next` lists, it does not advance. | New item R27 · recorded, 14.1 |
| Completion certificate (`eunomia/progress/certificate.py`) | Markdown close-out: what shipped, QA evidence, PR refs, residual risks, acceptance stamp. | Verify evidence is free text. | Fold into R9 · recorded, 14.1 |
| Human-resume dispatch burst (ADR-0031, `docs/v2/human-resume-jit-dispatch.md`) | A task whose question was just answered jumps the queue: four human-resumed dispatches before one normal one; `SKIP LOCKED` reservation with a transactional outbox. | No queue priority for answered work. | Fold into R1 · recorded, 14.1 |
| Lease-owned stage executions (ADR-0032, `eunomia/execution/leases.py`) | Execution key of task, stage and epoch; a key whose execution succeeded or was cancelled is returned without a new attempt, and a key under an unexpired running lease is refused; a queued, released or expired execution reacquires the same key and its attempt count rises (`leases.py`; its callers were not traced, 14.1); explicit `paused_hitl` state distinguishes waiting-on-a-human from lost. | Fencing tokens and lease TTLs; no stage epochs, no paused-on-human state. | Fold into R1 and R3 · recorded, 14.1 |
| Per-project pause, drain or immediate (ADR-0033, `eunomia/projects/pause.py`) | Pause one project's execution across restarts without stopping shared workers. | Automation switches are global. | New item R28 · recorded, 14.1 |
| Recovery and composer-retry surfaces (`api/routes/recovery.py`, `composer_retry.py`) | Named recovery verbs instead of ad-hoc SQL. | `./eve remote pending / resume / abort` cover interrupted API operations only. | With R25 · recorded, 14.1 |
| Atomic planner finalisation (ADR-0042) | Commit a running marker first, then artifact, review state and gate together in a fresh transaction, failure recorded separately. | Eve content writes documents and opens reviews as separate calls. | Rule for R6 · recorded, 14.1 |
| Task review, early PR, merge lifecycle (ADR-0026, `registries/lifecycle.yaml`) | Review a real diff; approving re-dispatches an idempotent publisher; squash merge policy as config. | Reviews attach to a commit SHA; no PR-first flow. | Fold into R8 and R11 · recorded, 14.1 |
| Pre-publish Git integration (ADR-0036, `agents/git_integration/agent.py`) | Merge fresh main (`git merge --no-ff`; the agent's prompt forbids rebase) in an isolated worktree; deterministic resolvers only for provable shapes (manifests, append-only docs, aggregators); a bounded agent for the rest, rejected if it strays or leaves markers; base SHA re-checked before push. | Every session merges `origin/main` by hand; `pnpm-lock.yaml` has a written recipe and nothing else does. | New item R29 · corrected, 14.1 |
| Governed review proposals (ADR-0029, `api/routes/prd_review.py`) | A comment never regenerates; a proposal bound to a component hash does, once accepted. | Document reviews carry findings; nothing distinguishes a remark from an authorised correction. | Note under R7 · recorded, 14.1 |
| Artifact lineage graph (ADR-0021, `SourceMaterialsPanel.tsx`) | A generated artifact can record its sources and their roles. ADR-0021 is Proposed, and whether every generated artifact records its sources is not established (14.1). | Document relations exist; generated presentations and docs do not record what fed them. | Fold into R21 · corrected, 14.1 |
| Thresholds registry (`registries/thresholds.yaml`) and HITL timeout sweep | One file for confidence floors, HITL timeout (24 h), supersede backstop, tool-call budget, max task estimate; a sweep that re-alerts on a HITL timeout and pauses nothing (it reads its timeout from `config.py`, `EUNOMIA_HITL_TIMEOUT_SECONDS`, not from this file). | Constants are spread across policies and code. | Fold into R1 and R3 · corrected, 14.1 |
| Per-agent model routing with fallbacks (`registries/models.yaml`; AUD-038 independent auditor never the engine that wrote the code) | An engine per role (several roles stay on the strongest model for quality), fallbacks for many roles but not the reviewer, QA or PRD composer, and the audit committee's model auditors routed away from the engine that authored the code (AUD-038; the AgentReviewer is not). | The OpenRouter route and the Codex lane are pinned in code and environment. | Fold into R14 · corrected, 14.1 |
| CLI failure taxonomy and per-worker account isolation (CLI-005, CLI-006) | `binary_missing`, `timeout`, `auth`, `rate_limit`, `nonzero_exit`, `parse_error`, `budget_exhausted`; `HOME` / `CODEX_HOME` per worker. | The Codex lane reports failures as text. | Fold into R14 · recorded, 14.1 |
| Project autonomy profile (`eunomia/autonomy/profile.py`) | Typed conventions agents read instead of asking; only high-risk questions reach a human. | `CLAUDE.md` and `.claude/rules` are prose. | Fold into R3 · recorded, 14.1 |
| Brownfield onboarding (ADR-0025, `eunomia/brownfield/`) | Feature inventory separate from the aspirational PRD; evidence rails drop claims whose paths do not resolve; targeted evidence search before declaring a feature missing. | Nothing comparable, and every Oshun root is brownfield. | Prerequisite for R6 · recorded, 14.1 |
| GitHub push webhook with self versus external distinction (`integrations/github_webhook.py`) | Keeps the system current when Elikem pushes from a laptop. | None. | Fold into R11 · recorded, 14.1 |
| Chatbot file generation (`file_gen.pdf`, `docx`, `pptx`, `xlsx`, `png`; `save_generated_file`) | Content-production verbs from chat, saved to the workspace. | Eve's turn accepts text only; presentation tools exist for slides. | Note under R21 · recorded, 14.1 |
| Steward memory and supervisory lane (`docs/v2/project-steward-memory-and-supervisory-lane.md`) | Semantic memory from approved components, episodic from applied decisions, retrieval capped and scoped, invalidated on artifact replacement; a second worker allow-listed to the Steward. | Operator memory is default-on; no decision-episode memory. | Fold into R3 · recorded, 14.1 |
| Architectural-layer dispatch phasing (`routing/layers.py`, `routing.yaml`) | "Finish the interface before wiring the backend" as scheduler policy, shipped off (`layer_phasing_enabled: false`). | None. | Note under R6 · corrected, 14.1 |
| Fleet resource snapshots and retained logs (ADR-0035, `observability/resources.py`, `PlatformLogsPage.tsx`) | Workers self-report memory and CPU; logs retained behind an authenticated proxy. | The 16 GB machine rules are prose; nothing measures them. | Fold into U5 · recorded, 14.1 |
| Non-destructive artifact integrity cron (`workspace/integrity.py`) | Content-hash reconciliation on a schedule. | Media restore verifies checksums; documents have no periodic integrity pass. | Later, minor · recorded, 14.1 |
| In-app documentation (`DocsPage.tsx`, `docsContent.ts`) | "When the lifecycle looks stuck", "Investigate before retrying". The guide's help search reads a catalogue of routes, not this content. | The Presentation Center teaches Eve; the copilot does not read it as help. | Fold into U6 · corrected, 14.1 |
| Operator mobile app (`mobile/`; the 2026-09-22 read named `eunomia/mobile/`) | Expo client that mirrors the web product; its project stack includes overview and reviews. | Admin mobile has no Eve function. | Fold into U10 · corrected, 14.1 |

### Additions to the build list

**R25 · Operations projection · NEXT · M.** One durable row per long
operation across renders, narration, builds, publications, watcher runs and
agent lanes: accepted, queued, running with real phases, terminal with a
sanitised reason; acknowledge, retry and cancel; a stream for U1 and U5; a
watchdog for abandoned rows; "what is Eve doing and why is it stuck" answered
from this table. From Eunomia `operations.py`. Where: BFF, workbench-kit
`JobQueueView`.

**R26 · Knowledge-gap queue from feedback · LATER · S–M.** Thumbs-down,
abstentions and refused actions become typed gap records grouped by screen
and skill, triaged into docs or tool tasks on the board. From Eunomia
assistant feedback: typed categories such as `no_knowledge` and `no_tool`,
grouped by screen, with triage statuses `documented` and `product_change`
(checked at `75f621cf`, section 14.1). Grouping by skill and filing board tasks are
this item's additions. Where: `feedback-store.ts`, R4 triage.

**R27 · Advance verb with a refusal matrix · NEXT · S–M.** `./eve advance
<family>` dispatches the one next step or refuses with a named reason: open
question, review pending, failed lane, no remote, no worker. From Eunomia
`continue_pipeline.py`. Where: CLI and worker API; the same verb backs U1's
"next move".

**R28 · Pause a family, drain or immediate · NEXT · S.** Durable per-family
pause that survives restarts; drain lets leases finish, immediate returns
them to ready. From Eunomia ADR-0033. Where: family policies, queue admission.

**R29 · Pre-publish Git integration · NOW · M.** Before any agent pushes:
rebase onto fresh main in an isolated worktree, resolve only provable shapes
deterministically (lockfile, append-only ledgers, generated boards), hand the
rest to a bounded agent that is rejected if it leaves markers or strays, run
the discovered gates, re-check the base SHA, then push. From Eunomia ADR-0036,
which merges fresh main rather than rebasing (section 14.1); rebasing is this
item's choice.
Where: a tool under `tools/`, called by `./eve done` and the automatic worker.
This repo's standing merge pain is the reason it is NOW.

### Folded into existing items

R1 gains the human-resume dispatch burst, lease stage epochs with a
paused-on-Elikem state and the HITL timeout sweep. R3 gains the autonomy
profile, Steward memory and the thresholds registry. R6 gains brownfield
intake as a prerequisite and the planner finalisation transaction rule. R8
and R11 gain the PR-first lifecycle and the push webhook. R9 gains the
completion certificate. R10 gains the weekly State of Affairs rollups and
export. R14 gains per-role routing with fallbacks and the CLI failure
taxonomy. R21 gains artifact lineage and file generation. U1 gains "next
move". U5 gains worker resource self-reporting. U6 gains the per-screen guide
and in-app help.

These folds rest on the Eunomia verification in section 14.1.

## 11. Zel second pass: excluded

The 2026-09-22 source-level sweep of Zel is excluded for the same reason as
section 4: Zel's source revision is unavailable for review (section 14), so
this analysis makes no claim about the features that sweep reported, and
nothing here implies that Zel lacks any of them. The build-list items this
section used to define are accounted for elsewhere: R25 to R29 come from
Eunomia and now close section 10; R30 is retained in section 6 on Eve's own
evidence; R31 and R32 are removed (section 15). The historical sweep is in
revision 14 of this document and in its Git history (see the header).

## 12. Second pass on Linear: the 2026 agent-era features in detail

The first pass read the changelog index and five docs pages. This pass read
each 2026 changelog entry, the docs for Linear Agent, Loops, Inbox, Reviews,
Releases, Insights, project graphs and Triage Intelligence, the developer
model for agent sessions, and the late-2025 entries that introduced Product
Intelligence and third-party agents. Rows repeat a section 5 feature only
when the detail changes what Eve should build.

| Linear feature | What it does, in detail | Eve today | Disposition |
| --- | --- | --- | --- |
| Loops mechanics (docs, 2026-07-20, 2026-09-14) | Triggers: schedule with timezone; issue events (status, priority, assignee, cycle, project, label); object events (project, initiative, cycle, release); a team's triage queue. Actions: comment, create issues and follow-ups, edit metadata, edit documents, post to Slack, start a coding session, call MCP connectors. Where the setting applies, a loop may write only to the entity that triggered it unless broadened; it does not apply to scheduled loops. Permissions per loop: team access, web access, code intelligence, coding sessions, externally synced issues, external sources. Draft-then-publish editing with every published version restorable. Run history with failures, and the agent can be asked to explain a run. A finished run can be continued as a conversation. Limits: eight conditions, eight actions. Runs pause when AI credits run out, except loops that only apply configured issue updates, which go on without them. | Watcher contract (ADR-0081) has sources, conditions, budgets and revocable authority; no write-scope default, no versioned drafts, no run history UI, no continue-as-conversation. | Fold into R5 |
| Skills (docs, 2026-06-04) | "Save this conversation as a skill": a successful chat becomes a personal skill invoked by slash command, auto-triggered when context matches, shareable, with templates for spec writing, weekly planning and drafting updates. | SMX task families are static and ratcheted; Voyager recipes can be saved but are not slash-invocable or auto-triggered; nothing turns a good Eve conversation into a skill in one step. | New item R33 |
| Workspace and personal guidance (Settings → AI) | Two behaviours on two pages. Linear Agent guidance is "a set of instructions for how Linear operates within your workspace and across teams", set workspace-wide, for a team or personal; workspace guidance "applies to everyone in the workspace by default", and guidance for one team can be written into the workspace guidance by naming that team. Triage Intelligence takes guidance at sub-team, parent-team or workspace level to refine its suggestions, and when several levels are set "all are considered but the most local guidance is considered first and weighted most heavily". Neither page says one level overrides another. | Operator memory holds facts; `CLAUDE.md` and `AGENTS.md` are prose for coding agents; there is no editable guidance document that Eve's runtime reads. | New item R34 |
| Agent sessions model (Linear for Agents 2025-05-20, developer docs) | An agent is a first-class user with a profile and activity page. Delegating an issue or mentioning the agent opens a session on that issue; the human stays the owner. Activities are typed: thought, action, elicitation (a question to the human), response, error; once a session is created, the agent must send an activity or update its external URL within ten seconds. Views filter by delegate; Insights segment by delegate; agents cannot sign in or administer. | Agents are scoped API identities; the assistant has a turn-event protocol; worker lanes write journal lines and a fleet read model, with no typed activity stream and no elicitation type. | New item R35 |
| Releases (2026-04-30, docs, 2026-06-18) | Pipelines, continuous or scheduled, owned by a team. A release is a commit SHA plus the issues auto-attached from commit references. Status automation: merged on PR merge, done when the release completes, so integrations fire when the change reaches customers. Release notes generated by the agent from included issues with a template; a changelog per pipeline; a GitHub Action and pipeline access keys; path filters for monorepos; frozen stages. | Publications carry freeze and promote for content; the Hetzner release controller deploys; the release-note feed exists; `done --commit` records a SHA. Code deploys to Dev and UAT are not objects that link board tasks. | New item R36 |
| Product Intelligence detail (2025-08-14, 2025-09-18, docs) | Suggestions for team, project, assignee, labels, related and duplicate issues with the reasoning on hover ("assignee on previous issues about performance in the mobile launch flow"); accept and dismiss feed the model; auto-apply per property or per value with auto-applied properties marked; one to four minutes per issue. | R4 as written. | Fold into R4 |
| Agent auto-fix on triage (2026-06-11) | On a new bug the agent gathers evidence through MCP (Sentry, Datadog), inspects code and attempts a fix before an engineer sees it; Linear reports roughly 30% of bugs resolved on the first pass. Scoping requests (investigate, plan) never trigger coding; only implementation does. | Crash groups and incidents are readable; nothing attempts a fix. | Fold into R4 and R18 |
| Reviews and Diffs (2026-05-28, 2026-07-30, docs, "Reviewing code in the agent era") | Review inside the tracker, synced both ways with GitHub. Reviews sidebar ordered by shipping proximity. Guided reviews group the change into sections that read like a story, core first, glue code separated, with an explanation each. Approve, request changes, merge. Iterate with a background agent from the diff and watch the diff update. Rich previews of images, PDFs and Markdown in the diff. Branch status widget (behind, conflicted). Preview-link shortcuts on the issue. Agent comments carry the agent's name, avatar and model. Providers can attach a risk score (Low, Typical, High, Very High) with an explanation. One click from the review to what inspired the change. | Reviews record findings against a commit and the reviewer's model; no diff surface, no risk score, no iterate-from-review. | Fold into R8 and R18 |
| Coding sessions detail (2026-08-20, docs) | Environments per repository: runtimes and versions, env vars, setup scripts, repository guidance. Model chosen per workspace. After coding the agent runs the app in a browser, walks the flow, captures before and after screenshots, fixes and reruns. Pricing shows tokens at provider rates plus sandbox blocks; spend caps per workspace and per user reset daily, weekly or monthly. Signed commits enforceable. Mobile Changes tab with line comments. | Automatic worker lane, cloud spend approvals, the Playwright inspect harness. No self-verification loop attached to the item, no per-lane caps with reset periods. | Fold into R14 and R18 |
| Priority Inbox mechanics (docs) | Priority tab auto-selected and customisable by source or filter; snooze and reminders with natural dates ("next quarter"); J and K navigation; quick search by title, id, type, team, project, priority; unsubscribe; a cap of 2,000 open notifications. | R2 as written. | Fold into R2 |
| Pulse (docs) | A feed of project and initiative updates, "for me", popular and recent; personal feeds from filters; a daily or weekly summary in the inbox at 06:00 local time; play the summary aloud. | Release-note feed; drawer voice can read a reply aloud. | Fold into R2 and R10 |
| LLM change summaries in notifications (2025-05-20) | Document changes summarised in inbox, email and Slack, filtering out grammar and style edits. | Content events record revisions; notifications carry no summary. | Fold into R2 and R7 |
| Advanced and AI filters, subscribe to a view (2026-02-13) | AND/OR groups; a natural-language filter; subscribe to a view to be notified when an issue starts matching; shared filtered views by URL. | `./eve list` filters. | Fold into R12 and R5 |
| Insights (docs) | Issue count, effort, cycle time, lead time, triage time and issue age as bars and scatterplots, sliced and segmented by label, project, team, assignee or status; burn-up; CSV export. | `./eve stats`. | Fold into R10 |
| Project graphs (docs) | Scope line, started and completed lines, weekly velocity, three completion estimates with a ±40% buffer, where remaining work sums the points of incomplete issues and an in-progress issue contributes one quarter of its points to that remaining work; refreshed hourly. | None. | Fold into R10, with the formula |
| Text attribution and agent-assisted editing (2026-07-23) | Author indicators tell a colleague's text from a loop's; agent edits highlight separately for review; restore to any checkpoint. | Revisions record the identity; no span-level attribution or highlight of agent edits. | Fold into R21 and U6 |
| Draft projects with the agent (2026-09-03) | "Create with Agent" composes a project brief from workspace context, auto-saving drafts. | Planning goals and plans exist; no agent-composed family from a brief. | Fold into R6 |
| Chat ergonomics (2026-04-23, docs) | Select text anywhere and press ⌘J to send it as context; chat as a maximised overlay; multiple chat tabs; browsable history. | Contextual invocation and page vision; the drawer has no history or tabs. | Fold into U6 |
| Team documents and anchored comments (2026-06-04) | Team pages that gather documents into sections; comments on a specific span or image of a description. | Threads exist; comments are not anchored to spans. | Fold into R21, minor |
| Web forms for Asks (2026-04-02) | Template-driven request forms; non-users can submit; follow-up by a synced email thread; every submission lands in triage. | Support queue and feedback store. | Fold into R17, minor |
| Desktop tabs with per-tab history and pinned tabs (2026-06-18) | Navigation that survives restarts. | None. | U9, minor |
| Team initiatives, private sub-teams, shared private issues, third-party app approvals, Okta-managed MCP, GitHub team reviews, Gemini Enterprise | Organisation features. | Not applicable. | SKIP |

The pages each row rests on (fetched on 2026-09-29; section 14):

- Loops mechanics: <https://linear.app/docs/loops>, <https://linear.app/changelog/2026-07-20-introducing-loops>, <https://linear.app/changelog/2026-09-14-loops-for-product-management>
- Skills: <https://linear.app/changelog/2026-06-04-team-documents>, <https://linear.app/docs/linear-agent>
- Workspace and personal guidance: <https://linear.app/docs/linear-agent>, <https://linear.app/docs/triage-intelligence>
- Agent sessions model: <https://linear.app/changelog/2025-05-20-linear-for-agents>, <https://linear.app/developers/agent-interaction>, <https://linear.app/docs/agents-in-linear>
- Releases: <https://linear.app/changelog/2026-04-30-releases>, <https://linear.app/changelog/2026-06-18-agent-assisted-project-updates>, <https://linear.app/docs/releases>
- Product Intelligence detail: <https://linear.app/changelog/2025-08-14-product-intelligence-technology-preview>, <https://linear.app/changelog/2025-09-19-auto-apply-triage-suggestions>, <https://linear.app/docs/triage-intelligence>
- Agent auto-fix on triage: <https://linear.app/changelog/2026-06-11-coding-sessions>
- Reviews and Diffs: <https://linear.app/changelog/2026-05-27-linear-diffs>, <https://linear.app/changelog/2026-07-30-coding-sessions-on-mobile>, <https://linear.app/docs/diffs>, <https://linear.app/now/reviewing-code-in-the-agent-era>
- Coding sessions detail: <https://linear.app/changelog/2026-08-20-coding-environments>, <https://linear.app/docs/coding-sessions>
- Priority Inbox mechanics: <https://linear.app/docs/inbox>
- Pulse: <https://linear.app/docs/pulse>
- LLM change summaries in notifications: <https://linear.app/changelog/2025-05-20-linear-for-agents>
- Advanced and AI filters, subscribe to a view: <https://linear.app/changelog/2026-02-13-advanced-filters-and-share-issues-in-private-teams>
- Insights: <https://linear.app/docs/insights>
- Project graphs: <https://linear.app/docs/project-graph>
- Text attribution and agent-assisted editing: <https://linear.app/changelog/2026-07-23-agent-assisted-editing>
- Draft projects with the agent: <https://linear.app/changelog/2026-09-03-priority-inbox>
- Chat ergonomics: <https://linear.app/changelog/2026-04-23-linear-agent-mcp-support>, <https://linear.app/docs/linear-agent>
- Team documents and anchored comments: <https://linear.app/changelog/2026-06-04-team-documents>
- Web forms for Asks: <https://linear.app/changelog/2026-04-02-web-forms-for-linear-asks>
- Desktop tabs with per-tab history and pinned tabs: <https://linear.app/changelog/2026-06-18-agent-assisted-project-updates>

### Additions to the build list

**R33 · Save a conversation as a skill · NEXT · M.** After a good exchange,
"save this as a skill" turns it into a named recipe with a slash command and
an auto-trigger description, stored under the ADR-0082 admission contract
beside the SMX families. Skills are versioned, listed in the drawer and the
Workspace, and never grant tools beyond the family's allowlist. From Linear
Skills. Where: `libs/oshun/assistant/src/skills`, the Voyager store, the
drawer composer.

**R34 · Standing guidance · NOW · S–M.** One editable guidance document for
Eve and one per family, versioned in Eve content, injected as the volatile
suffix of the prompt so the SMX prefix hash stays stable. Family
guidance overrides the global one where the two conflict: that is Eve's
design choice, not Linear's rule (Linear's Triage Intelligence considers
every level and weights the most local most heavily). This is where Elikem
writes "never suggest
duplicates for narration tasks" or "Isis work always cites a row in the
evidence file". From Linear's workspace, team and personal guidance. Where: Eve
content, `prompt-order.ts`, R3 policy.

**R35 · Agent session protocol · NEXT · M.** Every lane run on a task is a
session with typed activities: thought, action, elicitation (which is an R1
question), response, error. Eve's own choice: the first activity is a
thought, within seconds (Linear asks for any activity or an external URL
update within ten seconds). Sessions render in
the task journal, the fleet page and the attention queue; delegating to a
lane keeps Elikem as owner; views and metrics segment by lane. From Linear
for Agents. Where: `events` journal, `turn-event-protocol.ts` extended to
worker lanes, `tools/eve-reviewer/worker.mjs`, the automatic worker.

**R36 · Releases as objects · NEXT · M.** A release per Dev or UAT deploy:
commit SHA, tasks attached from commit trailers (R11), status shipped then
live, generated release notes into the existing feed, a changelog per
pipeline, and integrations that fire when the change is live rather than
merged. Content publications and code releases share one timeline. From
Linear Releases. Where: Hetzner release controller, publications, release
notes, `done --commit`.

### Folded into existing items

R2 gains snooze, reminders with natural dates, keyboard navigation, quick
search, a daily summary at a set hour read aloud through the drawer voice, and
change summaries in notifications. R4 gains the reasoning-on-hover, accept and
dismiss feedback, per-property auto-apply with marked properties, and an
attempt-a-fix step for crash groups before Elikem sees them. R5 gains
write-scope defaulting to the triggering entity (Eve's design choice, for scheduled runs too; Linear exempts scheduled loops), per-loop permissions,
draft-then-publish versions, run history that Eve can explain, and continuing
a run as a conversation. R6 gains "create a family from a brief". R8 gains
guided sections, a risk score with explanation, iterate-from-review that sends
a finding back to the lane, and the reviewer's model on every comment. R10
gains lead, cycle and triage time per lane, burn-up, CSV export and the
project-graph estimate formula. R12 gains AND/OR and natural-language filters
and subscribing to a view. R14 gains per-lane caps with daily, weekly and
monthly resets and a cost readout that shows tokens at provider rates. R18
gains environments per repository, the browser self-verification loop with
before and after screenshots, and the scoping-versus-implementation gate. R21
gains span attribution and anchored comments. U6 gains select-to-context,
tabs and history. U10 gains a Changes tab with line comments.

## 13. Graphs as Eve's spine

Elikem's prompt for this section: graphs and graph algorithms keep turning up
as the useful structure for knowledge bases, grounding, deterministic
representations and agentic systems; what can Eve get from that? The short
answer is that Eve is already half a graph system and does not yet use it as
one. The long answer follows: what exists, what the comparators do with
graphs, the principle, thirteen concrete capabilities with the algorithm each
one needs, and what not to build.

### 13.1 What Eve already has, verified in source

| Piece | Where | What it is |
| --- | --- | --- |
| Product graph | `libs/oshun/product-graph` | A typed property graph. Node kinds are a discriminated union (domains, flows, journeys, routes, docs pages, tours, work items, decisions, threads). Twelve edge kinds: `contains`, `follows`, `verifies`, `documents`, `anchors`, `tours`, `launches`, `visits`, `tracks`, `decides`, `discusses`, `supersedes`. Every edge carries provenance (source and content hash). The graph has a Merkle-style version; `validateProductGraph` fails CI on curation that points at unknown nodes; `indexProductGraph` builds adjacency; `whatTouchesRoute` is the first query. Compiled deterministically from the curated feature catalogue, the e2e journey inventory, the docs inventory and the route table. |
| Neighbourhood grounding | `apps/oshun/bff/src/workbench/graph-neighborhood.ts` | When a discussion anchors on a node, Eve's context is the neighbourhood: containment path and children, what `documents` it, what `verifies` it, routes reached by `visits`, work that `tracks` it. Every entry cites the edge kind that justified it. Per-relation cap 8, total cap 48, truncation counted. This is graph grounding done right, for one node kind family. |
| Board graph | `tools/eve-task-board` schemas | `task_deps` with kinds `depends-on` and `parent`, cycles refused, unresolved targets kept as `unresolved-dep` links rather than dropped. `task_links` to docs, code, tasks, slides, docs pages, work items. `doc_links` from documents to docs, code, tasks, slides, docs pages and URLs. `./eve context`, `docs related` and `docs stale` are all graph walks in disguise. |
| Eve task section of the product graph | `libs/oshun/product-graph/src/compilers/eve-tasks.ts`, `src/tools/build-artifact.ts` | The artifact's `todos` section projects the board into the product graph: `eve-family:` and `eve-task:` nodes, a family `contains` edge per task, and `tracks` edges from a task to the product nodes its text and curation name. It is built from the Eve task snapshot when the artifact build is given one (`options.eveTasks`, added 2026-09-21 in `df72d86b953`), otherwise from the TODOS tracker files. It is a projection inside the product graph, rebuilt with each artifact: it carries no dependency, lease or journal, and no document edge. |
| Retrieval | `apps/oshun/bff/src/assistant/retrieval` | BM25F and a 57,008 × 1,024 dense index fused by reciprocal rank fusion (k = 60). Lexical-only is the shipping default. There is no graph arm. |
| Explorer | `apps/oshun/web/src/app/assistant/graph` | Explore, work, decisions and changes tabs over the product graph with coverage counts. |
| Domain graph code | `libs/sophia/knowledge-graph`, `libs/sophia/citation-graph`, `libs/metis/knowledge-graph`, `libs/yemaya/canon-graph`, `libs/oshun/skill-system/src/dependencies.ts` | Traversal and algorithms, learning-path optimisation, co-citation and network analysis, temporal analysis, anomaly detection, construction, embeddings, GNN and temporal modules, canon consistency checks, a skill dependency resolver with cycle detection. Built for products, reusable by Eve. |
| Docs Center plan | Docs Center proposal (Eve document `58f43e8f-391d-8986-9093-899700919a63`, revision 1; the file `docs/proposals/DOCS_CENTER_PROPOSAL.md` at `10771d1e8e4`) §7 | A `systems/` entity catalogue: one code-linked node per domain, substrate, service, library, app, contract and UE module. |

Two gaps stand out immediately. The product graph, the board graph and the
document graph are three graphs with three schemas. The joins between them are
`task_links` and the product graph's Eve task section, which copies board
tasks into the product graph with their `tracks` edges to product nodes. Board
dependencies, leases and journals, and document sections, relations and
successors, are joined to nothing. And nothing computes over them: no ordering, no closure,
no diff, no metric. They are stored, walked one hop at a time, and drawn.

### 13.2 What the comparators do with graphs

- **Eunomia** is the most graph-native. The Sequencer alone owns the task DAG
  (D13); feature identity hashes (D16) are specified to keep node identity
  stable across regenerations (at `75f621cf` only Layer 1 hashing is built); the cascade (D3/D4) is a forward-closure walk from a changed
  scope; ADR-0021 (Proposed) lets an artifact record its sources as a lineage graph,
  optionally;
  brainstorm boards are typed component, edge and boundary contracts rendered
  to Excalidraw with stable element ids (BRD-009 to BRD-011); as-built
  architecture is derived from repository scans; sibling scope-glob overlap is
  named as a conflict code (F-OVERLAP), though the planner raises it as a
  clarification and the runtime overlap watchdog (SEQ-014) is open; the BR
  Auto-mapper ranks feature nodes by embedding. Checked at `75f621cf` in section
  14.1.
- **Linear** is a graph product without saying so: issue relations (blocks,
  blocked by, related, duplicate with merge), sub-issues, project to
  initiative, milestones, customer requests to issues, releases to issues via
  commit references. Triage Intelligence proposes `duplicate` and `related`
  edges from semantic similarity with reasoning. Loops fire on entity events
  and, where the setting applies (not to scheduled loops), may write only to
  the triggering entity unless broadened, which is a graph-locality rule for
  agent authority; G6 generalises it as Eve's own design choice. Code Intelligence joins the code
  graph to the work graph.

### 13.3 The principle

A model proposes; the graph disposes. Deterministic graph structure gives Eve
four things a retrieval index and a prompt cannot:

1. **Grounding with a citation.** An answer that names the edge it followed
   (`documents`, `verifies`, `tracks`) can be checked. Eve's neighbourhood
   packs already do this; the rest of Eve should.
2. **Computation instead of judgement.** What is blocked, what is stale, what
   a revision affects, what to do next, how far a family is from closure: all
   are graph queries with exact answers. Today several of these are model
   calls or Elikem's memory.
3. **Verifiable agent action.** Write scope as reachability, lineage as a
   backward closure, certification as a walk over evidence edges. An action
   outside the allowed subgraph is refused before a card is drawn.
4. **One map for Elikem.** Product, code, documents, tasks, decisions,
   releases, agents and evidence as one navigable structure instead of four
   apps.

The counter-principle matters as much: edges enter the graph only from
deterministic compilers (code, catalogues, links, commit trailers) or as
`proposed` with confidence and provenance until Elikem or the Steward accepts
them. A model may never write a fact edge directly. That is Eve's bright line
applied to structure.

### 13.4 Capabilities, with the algorithm each needs

**G1 · The Eve graph · NOW · M.** One typed property graph in PostgreSQL that
unifies the product graph, the board (families, tasks, dependencies, links),
Eve documents (sections, relations, successors), work items, decisions,
releases (R36), agent lanes and sessions (R35), and evidence artifacts.
The product graph's Eve task section already joins board tasks to product
nodes by `tracks` (section 13.1). What G1 adds is the rest of the join:
dependencies, leases, journals, documents, work items, decisions and releases
in one store, with one validation and one version.
Tables, not a graph engine: `graph_node (id, kind, attrs, content_hash)` and
`graph_edge (from, to, kind, provenance, valid_from, valid_to)`, recursive
CTEs for applicable walks over indexed PostgreSQL edges, preserving validity
semantics, and a Merkle version like the product graph has today. A separate
materialised adjacency cache is optional, justified by measured reader
performance (decided 2026-09-29 on the G01 review, question 3bbd2cd7). Compilers extend `@oshun/product-graph`'s pattern:
each source compiles to nodes and edges with a content hash, and a curation
target that points at nothing fails `./eve check`. A board link to a kind of
target no source compiles yet (today only slides, declared in the versioned
validation policy) is reported as unvalidated, never as resolved, until that
source exists (decided 2026-09-29 on the G01 Verify, question badfd8a3). Where:
`libs/oshun/product-graph` compilers, `tools/eve-task-board` schema, Eve
content relations.

**G2 · Grounding by neighbourhood everywhere · NEXT · M.** Extend the
neighbourhood pack from graph nodes to tasks, documents, families and
releases, so every Eve answer about work cites edges. Add a third retrieval
arm: candidates from a bounded walk out of the query's matched entities
(k-hop with edge-kind filters, or personalised PageRank seeded on the
matches) fused with BM25F and dense by the same reciprocal rank fusion.
Measure it with the existing relevance-set battery before it ships; the
hybrid audit's rule that a controlled invariant is not a relevance win applies.
Algorithms: k-hop with edge filters, personalised PageRank, three-arm RRF.
Where: `graph-neighborhood.ts`, `retrieval/hybrid-search.ts`.

**G3 · Impact and cascade as reachability · NEXT · M.** R7's engine. A
superseding document revision computes the forward closure over `anchors`,
`documents`, `tracks` and `derived-from` edges to the tasks, work items,
slides and artifacts it affects, marks them stale with the path as evidence,
and hands the set to the cascade. Stable identity comes from content hashes
plus an identity hash for regenerated nodes (Eunomia D16, specified there; only
Layer 1 hashing is built at `75f621cf`), so untouched nodes
survive a re-plan. Algorithms: forward closure with an edge whitelist,
snapshot diff (nodes and edges added, removed, rehashed). Where: content
events, `doc_links`, `task_links`.

**G4 · Scheduling on the DAG · NEXT · M.** `./eve next` becomes topological
order under priority, with the critical path to family closure as the ETA
input for R10, blocked-by chains rendered in U2 and U3, and scope overlap
between concurrently leased tasks refused as an edge conflict (named after
Eunomia's F-OVERLAP code; see 13.2) using declared path globs. Algorithms: Kahn's topological sort,
longest path in a DAG, glob interval overlap. Where: `lib/next.mjs`, the
board API, family policies.

**G5 · Proposed edges for triage · NEXT · M.** R4's inference layer.
`duplicates`, `related` and `caused-by` candidates from embedding nearest
neighbours, lexical overlap and shared neighbours in the graph (common
neighbours, Adamic-Adar), written as `proposed` edges with confidence and a
reasoning string, accepted by Elikem or the Steward under R3, and merged with
journals carried. Algorithms: kNN over the dense index, link prediction by
common neighbours, entity resolution on titles and paths. Where: the triage
route, `.cache/eve-task-board/knowledge.sqlite`.

**G6 · Write-scope locality · NOW · S.** A lane, loop or assistant turn may
write only to nodes within k hops of its anchor (the task it holds, the
watcher's source, the thread it was invoked from) unless the policy broadens
it. Checked as reachability before the confirmation card and refused with the
path that failed. Linear's loop rule, generalised. Algorithm: bounded
reachability. Where: workbench agent tools, watcher engine, R3 policy.

**G7 · Lineage for evidence · NEXT · M.** Every generated artifact (deck,
document, release note, screenshot, verify evidence, model output) records
`derived-from` edges to its inputs with content hashes. "Show everything that
fed X" is a backward closure; R9's certification walks it to prove a claim
rests on current inputs.
Eunomia ADR-0021. Algorithms: backward closure, hash-chain verification.
Where: publications, media, presentation sources, the evidence scripts.

**G8 · The code graph · NEXT · M–L.** Compile the monorepo's module, import,
route and ownership graph deterministically (tsconfig path mappings and
project manifests, since Nx's own graph breaks under worktrees here) and join
it to the product graph through `anchors` and `visits`. Then "what depends on
this library", "blast radius of this change", orphan modules and hotspots
(betweenness) are queries, and Code Intelligence questions are answered from
structure before a model reads a file. An orphan module is one that no
declared root reaches over import edges, where the roots are written down
per project: runtime entry points (each app's server and CLI mains),
framework-discovered files (Next.js `page.tsx`, `layout.tsx` and `route.ts`,
and the like), package `exports` and `bin` entries, and test and spec files.
In-degree alone is not the test: an entry point has no importer and is not
an orphan, while unused modules that import one another all have importers
and are orphans. R9's floor uses this definition. The compiler G08 has
shipped (`tools/eve-graph/compile-code.mjs`) marks a module orphan when it is
not an entry point and nothing imports it: that handles entry points and
misses unused modules that import one another, and `eve-gap-analysis:G08.5`
moves it to reachability.
Algorithms: reverse dependency closure, reachability from the declared roots,
betweenness centrality.
Where: a compiler beside `compileRouteSection`, `tools/eve-everywhere`
orphan and route probes.

**G9 · Family health as graph metrics · NEXT · S.** For R10, computed with no
model: frontier size (tasks whose predecessors are all done), unleased ready
count, longest blocked chain, stale documents on the frontier, verify edges
missing on shipped tasks. Where: family pages (U4), snapshots.

**G10 · Decision graph constraints · LATER · S–M.** Decisions already
`decides` nodes and `supersedes` each other. Give decisions a typed outcome,
require plan revisions to cite the decisions they rest on, and detect two
active decisions with conflicting outcomes on one node. Where:
`draft_decision`, `transition_decision`, R6.

**G11 · Shape rules in `./eve check` · NOW · S–M.** Extend the board check's questions (eight in `lib/check.mjs`, ten on the shared PostgreSQL board)
with graph shape rules the way `validateProductGraph` already validates
curation: a ready task links to at least one document or code path; a task
may not cite a superseded document without a successor; a shipped task has a
commit edge; a verify edge may not come from a failed journey; no `proposed`
edge older than a policy age without a decision. Algorithm: rule evaluation
over the graph, like SHACL shapes without the vocabulary. Where: `lib/check.mjs`.

**G12 · Temporal graph · LATER · M.** `valid_from` and `valid_to` on edges so
"what did the graph look like when this task shipped" and "what changed since
Elikem last looked" are queries. Feeds R2's change summaries, the explorer's
changes tab and R12's subscribed views. Algorithms: time-sliced queries,
snapshot diff. Where: G1's schema from the start, so no migration later.

**G13 · Graph views for Elikem · LATER · M.** The explorer draws
neighbourhoods; add a blocked-by chain view, a blast-radius view for a
document revision, and a lineage view for an artifact. Reuse
`@sophia/knowledge-graph` traversal and `@sophia/citation-graph` network
analysis rather than writing new code. Where: `GraphExplorerBoard`, U2, U3.

### 13.5 What not to build

- **No graph database engine.** At Eve's scale (thousands of nodes, tens of
  thousands of edges, 57 thousand retrieval chunks) PostgreSQL with recursive
  CTEs over indexed edges is enough (a materialised adjacency cache only when
  measured reader performance calls for one), and it keeps one backup,
  one credential and one audit path. Revisit only if a walk exceeds its
  reliability budget in the SLO contract.
- **No model-extracted knowledge graph as truth.** Entity and relation
  extraction from prose produces plausible edges, which is exactly what Eve
  must not treat as fact. Extraction may produce `proposed` edges only.
- **No embedding features in scheduling or checks.** Similarity belongs in
  retrieval and link prediction; ordering, closure and shape rules stay exact.
- **No reachability-as-permission.** The graph decides what an action may
  touch; the scoped credential decides whether the actor may act at all. Keep
  the two layers separate.

### 13.6 Order

G1 first, because everything else needs the unified schema, and G12's
validity columns go in with it. G6 and G11 next because they are small and
each one converts an existing rule from prose into a check. Then G3 and G4,
which give R7 and R10 their engines. G2 and G5 after the relevance battery
can measure them. G7 and G8 when R9 and R18 need them. G9, G10 and G13 ride
on the pages that show them.

## 14. Evidence boundary

- Eve inventory is from source in this checkout at commit `10771d1e8e4`:
  `tools/eve-task-board`, `infra/hetzner/eve`, `apps/oshun/eve-content`,
  `apps/oshun/bff/src/workbench` and `assistant`, `libs/oshun/assistant`,
  ADR-0080 to ADR-0088 (then files under `docs/adr/`, now Eve documents, each
  a relation of this document), and the Eve Everywhere initiative (Eve document `e5501406-96aa-8236-8d4d-1ca4e68c368c`, revision 1; the file `docs/agents/eve-everywhere-initiative.md` at `10771d1e8e4`). The
  `oshun-eve` MCP server failed to connect in the session that wrote this (its
  launcher could not find `node`), so no live board or document query was made.
  The 2026-09-29 revision was made with live queries: each retired file's Eve
  document read by ID (the relations of this document). It states no board
  totals (section 15).
- Eunomia: the 2026-09-22 analysis read `FEATURES.md`, `CLAUDE.md`,
  `README.md`, `docs/v2/agents-explained.md` and the ADR index through ADR-0047
  from an unpinned checkout, and executed nothing. It reported 506 checked and
  31 open items, "reconciled against code on 2026-08-29". Those counts match the
  file as committed at `4c397fac7e537d0245e472d9aefd429a043442b6` (2026-09-22T11:47:34Z, its last change on or
  before that day), whose ADR index also stops at ADR-0047. That is consistent
  with, not proof of, the commit the read saw. The 2026-08-29 note is the
  file's reconciliation record and counts 461 checked and 34 outstanding as of
  that date, so the 506 and 31 are later figures than the note.
  On 2026-09-29 the authoring worker checked every Eunomia claim in this
  document against `git@github.com:GreyChimp/eunomia.git` at `75f621cf59145167cbfc14055458260cb80a250f`
  (committed 2026-09-29T01:05:44Z), read-only, with excerpts: section 14.1.
  The repository is private and is not in the review package, so that check is
  the author's and is not independently verified; 14.1 gives each cited file's
  blob id so a reader with access can confirm it, and the cited text itself is
  copied into the evidence documents `96edcbf0-d40d-4fe5-a526-19dfd18e3bee` and `63fa87db-c65f-4c4b-93a3-9defa964ed81`. At that commit
  `FEATURES.md` has 510 checked and 32 open, the ADR index lists through
  ADR-0048 and `docs/v2/adr/` reaches ADR-0049. Of 72 claims, 57 are recorded,
  14 partly and 1 differs (these counts); the rows of sections 3 and 10, the
  Eunomia bullet of 13.2 and the build-list items that were partly supported
  are corrected in place.
- Zel: excluded. On 2026-09-29, from the Linux dev server, `git ls-remote` on
  `git@github.com:GreyChimp/zel-ai.git` and on the names `zel`, `zelai`,
  `Zel`, `zel-app`, `zel-platform` and `zel_ai` answered "Repository not found",
  and no Zel checkout exists on that host. This establishes only that the
  source could not be read here; it does not show that Zel lacks, or has, any
  capability. Sections 4 and 11 are exclusion notes; the historical analysis
  with the Zel material is revision 14 of this document (header).
- Linear is from public pages: the changelog through 2026-09-14
  (<https://linear.app/changelog>), "Introducing Linear Agent"
  (<https://linear.app/changelog/2026-03-24-introducing-linear-agent>),
  <https://linear.app/ai>, and the docs for triage, initiatives, customer
  requests, cycles and SLAs under <https://linear.app/docs>. Section 12 adds
  the individual 2026 entries (2026-02-05, 02-13, 04-02, 04-23, 04-30, 05-14,
  05-28, 06-04, 06-11, 06-18, 07-02, 07-20, 07-23, 07-30, 08-13, 08-20, 09-03,
  09-14), the 2025 entries for Linear for Agents (05-20), Product Intelligence
  (08-14) and auto-apply triage (09-18), the docs for Linear Agent, Loops,
  Inbox, Reviews (Diffs), Releases, Insights, project graphs, Pulse, Triage
  Intelligence and agents in Linear, the developer page for agent sessions,
  and the essay "Reviewing code in the agent era" on <https://linear.app/now>.
  Pricing and plan gates are as published there. A date is the one its page
  displays, not its URL's: auto-apply triage shows 2025-09-18 at a
  2025-09-19 address, and Linear Diffs 2026-05-28 at a 2026-05-27 address.
  Since the 2026-09-29
  revision each row of sections 5 and 12 names the pages it rests on. Each of
  those pages was fetched on 2026-09-29 and the detail the row relies on was
  found in its text, with one exception: the docs `editor` page did not carry
  document version history, so that row of section 5 cites the 2026-07-23
  entry (restore to a checkpoint) alone. Review 6d51a5df found that section
  12's guidance row did not match its pages (it said team guidance overrides
  workspace guidance, and quoted an example neither page carries); the row now
  quotes the two pages, fetched again on 2026-09-29.
- Section 10 comes from a second, source-level sweep on 2026-09-22 that
  walked Eunomia's module trees, route files, tool registries, YAML
  registries, web pages and planning documents; every row cites the path that
  proves the feature exists. Section 10's rows are checked at `75f621cf` in
  section 14.1 (one path, `eunomia/mobile/`, is the top-level `mobile/` there).
- Section 9 is from source and the showcase frames only; no server was booted
  and no page was rendered for this pass.
- Section 13's inventory of Eve's graph pieces is from source in this
  checkout (`libs/oshun/product-graph`, `graph-neighborhood.ts`, the board
  and document schemas, the retrieval directory, the four domain graph
  libraries). The comparator rows reuse sections 3, 5, 10 and 12. The
  algorithms named are standard and are not claimed to exist in Eve unless the
  row says so.
- Effort sizes are judgement against the existing seams named in each item,
  not measurements.

### 14.1 Eunomia claims checked at `75f621cf`

On 2026-09-29 the authoring worker (claude-board-worker) checked every
Eunomia claim in this document, read-only, against
`git@github.com:GreyChimp/eunomia.git` at `75f621cf59145167cbfc14055458260cb80a250f` (committed
2026-09-29T01:05:44Z; its tree is `4c3c6a1bb616ec5ee8ce30fa895c9670aa3f0ab8`). Nothing was executed.

**What this check is, and is not.** The Eunomia repository is private, so the
statuses and notes below are the author's findings. Everything they rest on is
copied into Eve evidence documents, all relations of this one and each pinned
at revision 1, so a reviewer without access can read it:

- the windows around every cited line (the cited line with three lines either
  side): `96edcbf0-d40d-4fe5-a526-19dfd18e3bee` for `FEATURES.md` and
  `63fa87db-c65f-4c4b-93a3-9defa964ed81` for the other 78 cited files and the
  `docs/v2/adr/` listing;
- `FEATURES.md` whole, every line numbered, in four parts: `56eb3103-0125-4145-a3e4-12e79e0d7484`, `cdf89cb5-bf8b-4ad9-84e8-62e1bcc4d8b2`, `da40a85b-c875-4942-92aa-4ce942aec830`, `5fe79165-5b7c-4996-8949-a2a6cd7856cf`. Each
  part counts its checked and open items, and the four add up to 510 and 32;
- every tracked path at the commit, in two parts: `9a5cbd8e-1149-403f-b26f-f8896e680d1b`, `7ad9fdf3-973a-4346-bd23-c148e935b57b`. A statement
  that a file exists, or does not, is checked there;
- the code behind the findings, in three parts: `0f7e3a47-ae6f-4b7c-b521-d788e1009bf1`, `676de162-a9a7-436f-8332-114fa9aa4659`, `baeebb4f-fe67-4855-a4d1-99a2ac11dc1d`: 38 numbered
  items, each a file whole, a range of its lines or a `git grep -n` over the
  commit, with its command. A note cites them as "code item N";
- a claim-to-source map, `10e3d44c-a5d7-401a-a4d1-f6e5e5759ee5`: each claim below with its status,
  its cited lines and the evidence document holding each, and its code items.

All 79 cited files' blob ids (listed at the end of this section) were checked
against the commit when the copies were made; a reader with access confirms a
file with `git rev-parse 75f621cf5914:<path>` and a line with
`git show 75f621cf5914:<path> | sed -n '<line>p'`. The copies are the
author's: they let a reviewer read what each finding rests on, and they do not
replace an independent fetch of the repository.

A claim is **recorded** when what the cited sources at that commit record
supports it: `FEATURES.md`, an ADR or other cited document, the cited lines,
the path list, or a code item. It is **partly** when they support some of it
(the row or passage is corrected in place and the note says what differs),
and **differs** when they contradict it. A statement about what Eunomia's
code does is made only where a code item or a cited line shows it; anything
wider (a caller not shown, a behaviour across the repository) is attributed
to what `FEATURES.md` or the ADRs record, or said to be not traced. No
Eunomia code was run and no Eunomia test was run; a test named here was read,
not run. Excerpts are quoted as `path:line` at that commit and cut at 160
characters. This baseline is not the unpinned checkout the 2026-09-22 analysis
read (section 14); each finding speaks for `75f621cf` only.

Totals, as the author found them: 72 claims; 57 recorded, 14 partly, 1 differs.


#### Section 3

- **§3 · ClarificationIssue**: recorded. ADR-0017 defines ClarificationIssue as the single uncertainty mechanism with typed scope, concern_kind, suggested_resolution and the open → answered → applied → resolved machine. eunomia/clarification/lifecycle.py lists the allowed transitions, rejects any other with F-STRUCT (assert_transition) and holds OPEN and ANSWERED issues as blocking their gate (blocks_gate) (code item 1). FEATURES.md records gate blocking (CI-014, count_blocking_for_gate) and scoped re-fire (CI-015, dispatch_scoped_refire_on_ci_resolved) as checked; their callers were not traced here. Caveats: ADR-0017 carves out DeveloperAgent (uses HITL gates), and FEATURES.md records four of the five non-dismiss resolution paths as open (CI-009..012), its own 2026-08-29 notes saying apply_resolution handles typed_answer and dismissed only. Excerpts: `docs/v2/adr/0017-clarification-issue.md:12` `` **One mechanism — `ClarificationIssue` — used by every agent that needs to surface uncertainty.** ``; `docs/v2/adr/0017-clarification-issue.md:32` `` Five resolution paths all converge on the same architectural mechanism: **workspace gets new content → emitting agent re-fires SCOPED to the issue's `scope.id`… ``; `eunomia/clarification/lifecycle.py:23` `_ALLOWED: dict[IssueState, frozenset[IssueState]] = {`.
- **§3 · Project Steward delegated authority**: partly. PST-001..010 and ADR-0040 support per-project presets, shadow mode, confidence threshold, mandatory high-risk escalation, a decision audit with policy snapshot, and Steward authority over gates (PST-005), task QA certification (PST-006), failed-task recovery (PST-007) and BugReport triage (PST-008). The audit is described as durable and idempotent, not immutable: the steward_decisions row carries a mutable state (queued/running/...) and attempt_history. Excerpts: `FEATURES.md:1441` `- [x] **PST-001** Per-project delegated-authority policy with presets, custom capabilities, shadow mode, confidence floor, and mandatory high-risk escalation`; `FEATURES.md:1442` `` - [x] **PST-002** Durable idempotent Steward decision audit (`steward_decisions`) with policy snapshot, model, rationale, confidence, risk flags, bounded retry… ``; `FEATURES.md:1447` `- [x] **PST-007** Failed-task recovery decisions with generation-safe retry or human escalation`.
- **§3 · Composer chain**: recorded. The EPL/STL/ATL checklists cover per-epic spec anchors and T-shirt estimates, story acceptance touchpoints and estimates, atomic-task scope_globs, dependency identification at each level, a sibling scope_glob overlap check, and ClarificationIssue emission at every planner. Acceptance criteria are carried as qa_criteria/acceptance_touchpoints, not a field literally named acceptance criteria. Excerpts: `FEATURES.md:1170` `- [x] **EPL-006** Per-epic T-shirt estimate (S / M / L / XL)`; `FEATURES.md:1203` `` - [x] **ATL-011** Sibling `scope_glob` overlap check (no two siblings can write the same files) ``; `eunomia/agents/atomic_planner/steps/granularity_and_overlap_check.py:8` `- ATL-011 (sibling overlap): for each story, compare every pair`.
- **§3 · Cascade barrier**: partly. The cascade part holds: SIGNAL marks scopes STALE and writes pause marks, SETTLE waits for worker acknowledgement and supersedes safe states (SEQ-010/011 checked; barrier.py), RECONCILE re-fires per-scope reconcilers and DISPATCH promotes ready tasks. The identity part is weaker: only D16 Layer 1 hashing is built (WS-009); identity_hash diffing across runs (WS-010) and the embedding two-layer match (WS-013) are unchecked, so the claim that identities survive a re-plan is the ADR's design rather than built behaviour. Excerpts: `FEATURES.md:1230` `- [x] **SEQ-011** SETTLE phase — waits for cooperative acknowledgement with timeout, enforces the blast-radius backstop, supersedes only safe states, and prese…`; `FEATURES.md:963` `` - [x] **WS-009** `identity_hash` computation (D16 Layer 1 — `sha256(slug(title) + "|" + sorted normalize(qa_criteria))`). Layer 2 (WS-013) still deferred. ``; `FEATURES.md:964` `` - [ ] **WS-010** `identity_hash` diff across runs (stable ID preservation + stale detection) — *not built (checked 2026-08-29): no identity_hash diffing across… ``.
- **§3 · WorkProduct**: recorded. ADR-0015 and WP-001..007 define WorkProduct as diff_set, scope_check, test_delta, decision_log, human_inputs and one free-form agent_notes slot, with decision_log read first by reviewers. Excerpts: `docs/v2/adr/0015-work-product.md:12` `**Structured-typed WorkProduct with one free-form slot.**`; `docs/v2/adr/0015-work-product.md:25` `` Reviewers always read `decision_log` first. `agent_notes` is the only field where the agent (or worker) can write prose. ``; `FEATURES.md:1272` `` - [x] **WP-005** `decision_log` field — list of `DecisionLogEntry { decision, rationale, at }`; read FIRST by AgentReviewer per ADR-0015 ``.
- **§3 · Certification committee**: recorded. ADR-0043 defines the three-tier committee (deterministic floor that gates alone, model ceiling with structural and security members, intent tier with feature and coherence auditors) and a novelty-terminated loop with a per-attempt budget; AUD-027 adds the whole-application render probe and AUD-040 the hybrid finding identity. Excerpts: `docs/v2/adr/0043-certification-requires-audit.md:56` `**Floor — deterministic, every sweep, no model calls, gates on its own.**`; `docs/v2/adr/0043-certification-requires-audit.md:85` `The sweep loop terminates on **novelty**, not on an iteration count: a sweep`; `FEATURES.md:1534` `- [x] **AUD-040** Hybrid finding identity — deterministic floor keys; lexical plus local-embedding candidate retrieval across members`.
- **§3 · BugReport pipeline and Auto-mapper**: partly. Five intake channels (BR-001), severity classification (BR-002), embedding rank (BRM-002), vision check (BRM-005) and confidence with top-3 alternatives (BRM-004) are all checked. The human-triager condition differs: ADR-0009 and the br_mapper README say a human triager always confirms before a task is derived (auto-conversion forbidden), not only when confidence is low; low confidence (F-LOW_CONF) only means the triager starts with a null mapping. Excerpts: `FEATURES.md:1404` `- [x] **BR-001** 5-channel BugReport intake (reporter portal · in-app widget · Slack webhook · Linear webhook · email)`; `FEATURES.md:1419` `- [x] **BRM-005** Vision check on evidence (screenshots / video boosts or lowers confidence)`; `docs/v2/adr/0009-external-input-loop.md:15` `Severity check + human triager confirm before a task is derived (auto-conversion is explicitly forbidden — see Phase 2 brainstorm).`.
- **§3 · Progress snapshots and burn-down**: recorded. ADR-0022 and PG-001..015 cover append-only snapshots triggered by cron, cascade DISPATCH and gate state changes, velocity and ETA with low/medium/high confidence tiers, and a sparkline on project list cards. Excerpts: `FEATURES.md:1471` `` - [x] **PG-001** `project_progress_snapshots` table + Alembic migration (append-only, typed `cause`) ``; `FEATURES.md:1478` `- [x] **PG-008** Velocity + ETA computation with confidence tiers (low <3d / medium 3-14d / high 14d+)`; `FEATURES.md:1482` `- [x] **PG-012** Project list card sparkline integration`.
- **§3 · LLM control plane**: recorded. CLI-007 (durable usage ledger), CLI-008 (organization/project/workload allowances with periods, budget-aware routing and hard stops), LLM-CP-002 (account health, cooldowns and fallback chains) and AUD-031 (prompt version and downstream outcome on llm_invocations) are all checked; code lives under eunomia/llm_control/. Excerpts: `FEATURES.md:805` `- [x] **CLI-008** Scoped allowances — revisioned organization/project/workload token and cost policies with daily/weekly/monthly/lifetime periods, percentage a…`; `FEATURES.md:23` `- [x] **LLM-CP-002** — CLI account health, cooldowns, and policy fallback chains.`; `FEATURES.md:1525` `` - [x] **AUD-031** Prompt version + downstream outcome on `llm_invocations`; per-version evaluation ``.
- **§3 · Governed agent configuration**: recorded. ADR-0045 makes agent revisions pin exact promoted skill versions and treats prompt versions as immutable candidates promoted after evaluation; AGC-002, SKL-001 and PRM-001 are checked, including rollback. Excerpts: `docs/v2/adr/0045-governed-agent-configuration.md:25` `behavior contract. Active agent revisions pin exact promoted skill versions;`; `FEATURES.md:32` `- [x] **AGC-002** — Immutable agent revisions with optimistic concurrency and activation checks.`; `FEATURES.md:42` `- [x] **PRM-001** — Global and project prompt lineage with isolated overrides, evaluation, promotion, and rollback.`.
- **§3 · Master/worker distributed execution**: recorded. FEATURES.md records WRK-003/005/006/009/010 as checked: Redis Streams consumption, heartbeats, the stale-heartbeat watchdog, the bare-clone cache with worktrees and async push after every commit-emitting step. Stale pending jobs are reclaimed through XAUTOCLAIM in the worker's claim_loop.py (code item 3), and the WRK-006 watchdog's docstring says it marks stale workers offline (code item 4). Excerpts: `FEATURES.md:813` `` - [x] **WRK-003** Worker entrypoint: `eunomia.workers.consumer` subscribes to `eunomia:jobs:atomic` via Redis Streams XREADGROUP ``; `FEATURES.md:819` `` - [x] **WRK-009** Bare-clone cache at `~/repos/<project>.git/`; lightweight worktrees share objects ``; `eunomia/workers/claim_loop.py:1059` `"""Try to reclaim one stale Redis pending entry via XAUTOCLAIM.`.
- **§3 · Environment sessions**: recorded. ESP-017 covers the EC2 code-server human IDE path (ESP-015 the code-server runtime), and ESP-019 covers artifact-only evidence workspaces with signed artifact URLs, all under ADR-0024. Excerpts: `FEATURES.md:1393` `` - [x] **ESP-017** EC2 human IDE production path — `aws_ec2_ide` providers launch one tagged task VM per human session, clone the repo, start code-server on tha… ``; `FEATURES.md:1395` `- [x] **ESP-019** Artifact-only evidence workspaces + signed artifact URLs`.
- **§3 · Notes, brainstorm boards, readiness signal**: recorded. NOT-001..005 implement draft → promote notes; BRD-009..011 add typed semantic contracts rendered to Excalidraw and agent-owned generated architecture boards; RDY-002/004 compute a green/amber/red readiness signal surfaced to PRDComposer as context. RDY-003 (thin-workspace ClarificationIssue) is unchecked, but the claim does not depend on it. Excerpts: `FEATURES.md:986` `` - [x] **NOT-003** `POST /v1/notes/{id}/promote` — writes `notes/{slug}-{shortid}.md`, registers `Artifact(type=NOTE)`, links back via `promoted_artifact_id` ``; `FEATURES.md:1003` `` - [x] **BRD-009** Typed `ArchitectureModel` + `DiagramSpec` semantic JSON contracts with validated component, edge, boundary, evidence, uncertainty, and source… ``; `FEATURES.md:1115` `` - [x] **RDY-002** ReadinessSignal computation (🟢🟡🔴) — five hardcoded rules (`brief_present`, `notes_present`, `research_present`, `interviews_present`, `recent… ``.
- **§3 · Work intake**: recorded. ADR-0038 and WRQ-010 describe AUTO choosing validated direct, additive for an existing PRD, or full composition; WRQ-009 makes duplicate detection advisory and non-suppressing. Excerpts: `docs/v2/adr/0038-post-creation-work-intake.md:35` `` `AUTO` deterministically chooses validated direct when that complete contract ``; `FEATURES.md:1359` `- [x] **WRQ-009** Advisory embedding/lexical duplicate detection requires explicit confirmation and never suppresses work`; `FEATURES.md:1360` `- [x] **WRQ-010** Deterministic automatic selection chooses validated direct, additive for an existing PRD, or full composition for an initial project`.
- **§3 · Chatbot with mutation gating**: recorded. ADR-0011 states read tools execute immediately and mutation tools need a confirmation step; CB-002/CB-003 record the executor and mutation_gate that implement it. Excerpts: `docs/v2/adr/0011-chatbot.md:16` `- Read tools execute immediately; mutation tools surface a confirmation step ("about to approve Gate 2 — confirm?") before executing.`; `FEATURES.md:1427` `` - [x] **CB-003** Mutation gating (any state-changing tool requires confirmation) *(`chatbot/mutation_gate.py` + `POST /chatbot/tool-calls/{id}/confirm`; UI con… ``.
- **§3 · F-classified failures**: recorded. CLAUDE.md's 'No Silent Failures' rule requires every failure to be F-classified and forbids returning None without a code; eunomia/core/errors.py's docstring calls it the F-classification registry, each code mapped to HTTP. Excerpts: `CLAUDE.md:97` `Every failure is F-classified. Silent failures are bugs.`; `CLAUDE.md:111` `` Every code path that can fail must classify the failure. Returning `None` without a code is forbidden. ``; `eunomia/core/errors.py:1` `"""F-classification registry. Every failure carries a stable code mapped to HTTP."""`.
- **§3 · QA Playbook with two independent testers**: recorded. ADR-0047 requires two independent testers, with results hidden until quorum (blind pass) and per-scenario results; QAP-004/006 are checked, and tester and lead invitations appear in the 2026-09-14 QAP status note and the QA playbook runbook (Workforce → Invite). Excerpts: `docs/v2/adr/0047-qa-playbook-human-qa.md:9` `it inside Eunomia immediately. The playbook's non-negotiable rules are: two`; `FEATURES.md:1318` `- [x] **QAP-004** Independent testing: no re-claiming reviewed work, results hidden from other testers until quorum, leads and admins see all, split verdicts r…`; `FEATURES.md:400` `> Staging fixes: tester and lead invitations, a scoped QA plan preflight, and`.
- **§3 · Design Studio parity**: recorded. ADR-0046 targets parity with Claude Design, and DSG-034 (share links), DSG-035 (PDF/PNG export) and DSG-036 (organisation design-system library) are checked. The ADR's own status line still reads Proposed. Excerpts: `docs/v2/adr/0046-design-studio-parity.md:1` `# ADR-0046 — Design Studio parity with Claude Design`; `FEATURES.md:1054` `` - [x] **DSG-035** PDF and PNG export: `export_render` worker job drives Chromium over CDP, stores the file in the artifact object store as a `design_export` ar… ``; `FEATURES.md:1055` `- [x] **DSG-036** Organisation design-system library: publish an approved project system or upload tokens into immutable object storage`.
- **§3 · Identity, organisations, reviewer roles**: recorded. The AU section (D20 / ADR-0020) implements users, orgs and org memberships, project participant roles, a reviewer pool and magic links; the feature label's 'reviewer roles' maps to AU-010..013. Excerpts: `FEATURES.md:852` `` - [x] **AU-001** `users` table + Alembic migration ``; `FEATURES.md:858` `` - [x] **AU-004** `project_memberships` table + Alembic migration (`project_id`, `user_id`, `participant_role ∈ {product_owner, triager, editor, viewer}` ``; `FEATURES.md:875` `` - [x] **AU-015** `magic_links` table + Alembic migration ``.
- **§3 · introduction**: recorded. CLAUDE.md names the agent chain (PRDComposer through AgentReviewer, plus QAAgent, ReadinessAgent and BR Auto-mapper) and humans approving at four gates; README and CLAUDE.md define master and worker roles, CLAUDE.md makes LLM access CLI-only through Claude and Codex (ADR-0018), and ADR-0043 adds the certification committee. The /mnt/volume300 path is the author's local checkout and was not checked. A per-task qa_gate exists besides the four planning gates. Excerpts: `CLAUDE.md:5` `with humans approving at four gates (Gate 1: PRD, Gate 2: tech spec, G3a: epics, G3b: stories)`; `CLAUDE.md:113` `## Critical Rule: LLM Access Is CLI-Only — No SDK, No API`; `README.md:24` `` Eunomia v2 runs in **two roles** — master and worker — set by `EUNOMIA_ROLE`. ``.

#### Section 10

- **§10 · Operations projection**: recorded. All four cited files exist (tree listing). The operations inventory (docs/v2/operations-inventory.md) requires creation before durable dispatch, real phases and heartbeats with no synthesized percentages, sanitized terminal reasons and a watchdog for abandoned rows. The routes include list, detail, acknowledge, retry, cancel and two SSE streams (code item 5), and workers/watchdog.py calls fail_stale_operations (code item 6). Excerpts: `docs/v2/operations-inventory.md:28` `- Workers report real phases and heartbeats. They never synthesize percentages`; `docs/v2/operations-inventory.md:32` `- The watchdog fails abandoned operations after the configured stale window.`; `eunomia/api/routes/operations.py:277` `@router.post("/v1/operations/{operation_id}/cancel", response_model=OperationResponse)`.
- **§10 · Contextual guide tools**: recorded. eunomia/chatbot/registry.py declares the four READ tools (screen explanation from a route key, product-knowledge search, sourced next steps, goal-to-route destination) (code item 7); eunomia/product_catalog.py's docstring calls it the product knowledge the contextual guide uses; web/src/components/guide holds EunomiaGuide.tsx and guideContext.ts (tree listing). Excerpts: `eunomia/chatbot/registry.py:71` `description="Explain a Eunomia screen from its safe route key and available actions.",`; `eunomia/chatbot/registry.py:93` `description="Return approved, sourced next steps for a known Eunomia screen or project workflow.`; `eunomia/product_catalog.py:1` `"""Permission-neutral product knowledge used by the contextual guide."""`.
- **§10 · "What is Eunomia doing / why is it stuck"**: recorded. get_active_operations, get_operation_details and explain_blocker are READ tools in chatbot/registry.py, and get_active_operations is described for 'what Eunomia is doing' questions (code item 7). Excerpts: `eunomia/chatbot/registry.py:110` `description="Get truthful persisted state and phase history for one operation.",`; `eunomia/chatbot/registry.py:121` `description="Explain failed or waiting operations and their recovery path.",`; `eunomia/chatbot/registry.py:152` `"and work needing attention. Use when the user asks what Eunomia is doing, "`.
- **§10 · Assistant knowledge-gap records**: recorded. eunomia/api/routes/assistant_feedback.py has the six failure categories and a route_key on each gap (code item 38); the guide's cited lines show its ThumbsDown control, and AssistantInsightsSettingsPage.tsx describes itself as grouping unresolved questions by frequency and screen. Excerpts: `eunomia/api/routes/assistant_feedback.py:24` `"no_knowledge", "no_tool", "permission", "ambiguous", "runtime_failure", "unhelpful"`; `web/src/pages/AssistantInsightsSettingsPage.tsx:39` `description="Questions users could not resolve, grouped by frequency and the screen where confusion occurred."`; `web/src/components/guide/EunomiaGuide.tsx:607` `{recorded ? <Check className="h-3 w-3" /> : <ThumbsDown className="h-3 w-3" />}`.
- **§10 · State of Affairs weekly report**: recorded. eunomia/state_of_affairs/ (report.py, html_export.py) and web/src/pages/StateOfAffairsPage.tsx exist (tree listing). The per-project schema carries progress_delta, eta_confidence, gates_pending, bug_reports_open, risk_flags and insight, and the report adds biggest_movers, stalled_projects and timeline_watchlist (code item 9); the routes are the report, export.html and POST /v1/state-of-affairs/snapshots (code item 8). Whether anything schedules a weekly capture was not traced here. Excerpts: `eunomia/schemas/state_of_affairs.py:32` `progress_delta: int | None = None`; `eunomia/schemas/state_of_affairs.py:87` `biggest_movers: list[StateOfAffairsProject]`; `eunomia/api/routes/state_of_affairs.py:113` `"/v1/state-of-affairs/export.html",`.
- **§10 · Dashboard "Suggested next move", "How to avoid stalls", "Due soon"**: partly. web/src/pages/DashboardPage.tsx has the 'Suggested next move', 'How to avoid stalls' and 'Due soon' (projects with targets in the next 14 days) panels. The suggested next move is not a computed single item: it is one of two fixed sentences, chosen by whether any active project exists, and the stall coaching is fixed copy (code item 22). Excerpts: `web/src/pages/DashboardPage.tsx:279` `title="Suggested next move"`; `web/src/pages/DashboardPage.tsx:283` `: "Open a recent project and check Pipeline, Inbox, and Workspace before starting new work."`; `web/src/pages/DashboardPage.tsx:291` `body="Keep source material specific, answer clarification issues quickly, and approve gates only after reviewing the artifact."`.
- **§10 · Continue-pipeline verb with a refusal matrix**: recorded. eunomia/lifecycle/continue_pipeline.py's docstring describes a refusal matrix (open clarifications, rejected/pending planning gates, failed tasks, missing git remote, no worker) and dispatching the one unstarted step (initial PRDComposer or a failed planner re-run); its cited line 223 is the missing_git_remote refusal. Excerpts: `eunomia/lifecycle/continue_pipeline.py:3` `Reads the live phase + gates + planner state, applies a refusal matrix (open`; `eunomia/lifecycle/continue_pipeline.py:4` `clarifications, rejected/pending gates, failed tasks, missing git remote, no`; `eunomia/lifecycle/continue_pipeline.py:5` `worker), and dispatches the one clearly-unstarted next step — the initial`.
- **§10 · Completion certificate**: recorded. eunomia/progress/certificate.py assembles and renders to markdown a certificate with shipped work and PR/commit refs, certification and QA-gate counts, open risks and the completion/acceptance stamp. Its 'QA evidence' is the certified-task and approved-QA-gate counts, not per-scenario detail (code item 10). Excerpts: `eunomia/progress/certificate.py:3` `` `complete_project` stamps a DB blob; this assembles the human-facing certificate: ``; `eunomia/progress/certificate.py:4` `what shipped, the QA evidence behind it, PR/commit references, any residual`; `eunomia/progress/certificate.py:174` `lines.append("## Open risks")`.
- **§10 · Human-resume dispatch burst**: recorded. ADR-0031 and docs/v2/human-resume-jit-dispatch.md give human-resumed tasks a scheduling class with a burst of four before one normal task, and reserve READY rows with FOR UPDATE SKIP LOCKED plus a transactional outbox row in the same commit. Excerpts: `docs/v2/human-resume-jit-dispatch.md:23` `used. While both classes contain work, the scheduler emits at most four human resumes`; `docs/v2/human-resume-jit-dispatch.md:49` `` Reservation locks READY rows with `FOR UPDATE SKIP LOCKED`, stamps `dispatched_at`, ``; `docs/v2/adr/0031-postgres-authoritative-jit-dispatch.md:24` `` uses `normal`. The scheduler permits a burst of four human-resumed tasks and then ``.
- **§10 · Lease-owned stage executions**: recorded. ADR-0032 keys every task-scoped run by task, stage and a monotonic epoch, makes completed keys suppress duplicate deliveries, and uses an explicit paused_hitl human-wait state. eunomia/execution/leases.py computes the task:stage:epoch key. acquire_execution_lease returns already_succeeded or already_cancelled without a new attempt for a finished key (lines 115-119) and refuses a key whose running lease has not expired (lease_active, lines 121-128); any other existing row, queued after a release or an expired lease, is leased again and its attempt count rises (line 146) (code item 2). Its callers were not traced here. Excerpts: `docs/v2/adr/0032-durable-task-stage-executions.md:16` `` `task_execution_attempts` row. Its execution key combines task, stage, and a ``; `docs/v2/adr/0032-durable-task-stage-executions.md:27` `` Reviewer rework is an explicit human-wait state (`paused_hitl`), not `built`. ``; `eunomia/execution/leases.py:1` `"""PostgreSQL-authoritative execution leases for task-scoped agent jobs."""`.
- **§10 · Per-project pause, drain or immediate**: recorded. ADR-0033 defines a durable per-project pause with drain and immediate modes that survives restarts and does not reduce service for other projects; eunomia/projects/pause.py reads the pause state from the Project row (code item 11). Excerpts: `docs/v2/adr/0033-project-execution-pause.md:17` `` `drain` prevents new handlers from starting while allowing current handlers to ``; `docs/v2/adr/0033-project-execution-pause.md:33` `- Pausing one project does not reduce service for other projects.`; `eunomia/projects/pause.py:1` `"""Authoritative project execution-pause reads."""`.
- **§10 · Recovery and composer-retry surfaces**: recorded. eunomia/api/routes/recovery.py exposes POST /v1/projects/{id}/recovery/reconcile, the idempotent stranded-work reconciler, behind a user principal and assert_can_admin_project (code item 12), and composer_retry.py exposes POST /v1/projects/{id}/composer/{name}/retry. That is two named verbs; the files do not state the 'instead of ad-hoc SQL' rationale. Excerpts: `eunomia/api/routes/recovery.py:1` `"""Authenticated operator recovery surface for stranded durable work."""`; `eunomia/api/routes/recovery.py:31` `summary="Run the idempotent stranded-work reconciler for one project",`; `eunomia/api/routes/composer_retry.py:196` `"/v1/projects/{project_id}/composer/{composer_name}/retry",`.
- **§10 · Atomic planner finalisation**: recorded. ADR-0042 sets three transaction boundaries: commit the running marker first, commit artifact, review state, audit evidence and gate handoff together in a fresh transaction only on success, and record failure in a separate short transaction. Excerpts: `docs/v2/adr/0042-atomic-planner-finalization.md:23` `` 1. Commit the durable planner `running` marker before workspace hydration or ``; `docs/v2/adr/0042-atomic-planner-finalization.md:25` `2. Execute the agent in a fresh transaction. Commit the artifact, structured`; `docs/v2/adr/0042-atomic-planner-finalization.md:28` `3. Record a returned or escaped failure in a separate short transaction.`.
- **§10 · Task review, early PR, merge lifecycle**: recorded. ADR-0026 opens the PR early so the human reviews the real diff, and approving a qa_gate with no PR re-dispatches the idempotent publisher. eunomia/registries/lifecycle.yaml (the cited registries/ path is under eunomia/) sets the lifecycle flags with merge_method: squash. Excerpts: `docs/v2/adr/0026-task-review-pr-merge-lifecycle.md:19` `> **re-dispatches the idempotent publisher** and asks the operator to review once it appears,`; `eunomia/registries/lifecycle.yaml:3` `#   pr_publish_on_review_accept  (#1) open the PR when the AgentReviewer`; `eunomia/registries/lifecycle.yaml:23` `merge_method: squash`.
- **§10 · Pre-publish Git integration**: partly. ADR-0036 describes an isolated worktree against the freshly fetched default branch, deterministic resolvers only for manifests, append-oriented docs and aggregators, a scope-bounded GitIntegrationAgent rejected for out-of-scope files or leftover markers, and a base-SHA re-check before push. It is a merge, not a rebase: workers/git_integration.py runs `git merge --no-ff --no-edit <base_sha>` in the worktree (code item 23), and the agent's rules forbid it to rebase or merge itself (code item 24). Excerpts: `docs/v2/adr/0036-pre-publish-git-integration.md:20` `1. Deterministic resolvers handle only provable merge shapes: structured package`; `docs/v2/adr/0036-pre-publish-git-integration.md:27` `4. The base SHA is checked again immediately before push. A moved base retries`; `eunomia/workers/git_integration.py:179` `"merge",`.
- **§10 · Governed review proposals**: recorded. ADR-0029 says a normal review comment never regenerates an artifact, proposals bind to an immutable artifact, component and base content hash, and only an accepted proposal is dispatched to a composer; eunomia/api/routes/prd_review.py has the proposal, vote and apply routes (code item 13). The ADR's status line reads 'Implemented locally (uncommitted)'. Excerpts: `docs/v2/adr/0029-governed-review-proposals.md:8` `Review feedback and agent uncertainty are separate concepts. A normal review comment`; `docs/v2/adr/0029-governed-review-proposals.md:27` `1. Proposals bind to an immutable artifact, component, and base content hash.`; `eunomia/api/routes/prd_review.py:719` `@router.post("/v1/projects/{project_id}/prd-review/proposals/{proposal_id}/apply")`.
- **§10 · Artifact lineage graph**: partly. ADR-0021 specifies the artifact_sources graph with typed roles, and its status is Proposed. The ArtifactSource table and GET /v1/artifacts/{artifact_id}/sources exist (code item 25), and web/src/components/inbox/SourceMaterialsPanel.tsx shows the role labels. That every generated artifact records its sources is not established: the ADR is only proposed, and which callers declare sources was not traced here. Excerpts: `docs/v2/adr/0021-artifact-lineage.md:16` `` **Introduce a typed, persisted artifact-lineage graph: every Artifact produced by an agent carries one or more `ArtifactSource` rows ``; `eunomia/workspace/artifacts.py:98` `` edges that fed it. `sources` is optional; agents that don't yet ``; `web/src/components/inbox/SourceMaterialsPanel.tsx:13` `primary_input: "Primary input",`.
- **§10 · Thresholds registry**: partly. eunomia/registries/thresholds.yaml holds confidence floors, hitl_timeout_seconds 86400 (24h), the cascade supersede_backstop_count, the tool_call_budget and atomic_planner.max_task_estimate_hours (code item 27). The timeout sweep only alerts: clarification/hitl_timeout.py describes itself as read-only, re-emitting hitl_timed_out events and leaving the clarification and its task as they are, so it pauses nothing (code item 26). It takes its threshold as an argument, typically settings.hitl_timeout_seconds from config.py (code items 26 and 28), not from thresholds.yaml. Excerpts: `eunomia/registries/thresholds.yaml:10` `hitl_timeout_seconds: 86400              # 24h before alerting operator`; `eunomia/registries/thresholds.yaml:52` `supersede_backstop_count: 20              # pause for operator above this blast radius`; `eunomia/clarification/hitl_timeout.py:9` `This is **not** a hard fail per CLAUDE.md:`.
- **§10 · Per-agent model routing with fallbacks**: partly. eunomia/registries/models.yaml routes each role to an engine, with *_fallback entries for many roles (developer, audit members, steward, git_integration, design and QA-plan composers) (code item 29). Three parts go further than the source: (1) the policy is not 'cheapest per role' (several roles stay on opus on quality grounds, per models.yaml); (2) not every role has a fallback (reviewer, qa and prd_composer have none in models.yaml); (3) author-engine exclusion applies to the four audit.* committee workloads only (_AUDIT_MODEL_WORKLOADS in llm_control/service.py, code item 30), not the AgentReviewer. Excerpts: `eunomia/registries/models.yaml:35` `developer_fallback:  claude-code`; `eunomia/llm_control/service.py:304` `"""Route model auditors away from the engine that authored accepted code."""`; `eunomia/llm_control/service.py:305` `if not project_id or decision.workload_key not in _AUDIT_MODEL_WORKLOADS:`.
- **§10 · CLI failure taxonomy and per-worker account isolation**: recorded. FEATURES.md records CLI-005 (HOME / CODEX_HOME per-worker account override) and CLI-006 (the seven F-CLI_FAIL classes) as checked; eunomia/core/llm/engines.py's cited line sets CODEX_HOME, and cli_router.py raises binary_missing (code item 14). Excerpts: `FEATURES.md:802` `` - [x] **CLI-005** Per-worker CLI account isolation — `HOME` override for Claude (reads `~/.claude/`), `CODEX_HOME` override for Codex ``; `FEATURES.md:803` `` - [x] **CLI-006** F-CLI_FAIL classification: `binary_missing` · `timeout` · `auth` · `rate_limit` · `nonzero_exit` · `parse_error` · `budget_exhausted` ``; `eunomia/core/llm/engines.py:293` `env["CODEX_HOME"] = self.account_dir`.
- **§10 · Project autonomy profile**: recorded. eunomia/autonomy/profile.py defines a typed ProjectAutonomyProfile of conventions (web framework, ORM, CI, dependency and GitHub CLI policies) (code item 15); its docstring says defaults decide low-risk choices while high-risk product, security, schema and scope questions still go to HITL. Excerpts: `eunomia/autonomy/profile.py:4` `surface: project/org defaults decide low-risk implementation choices, while`; `eunomia/autonomy/profile.py:5` `high-risk product, security, schema, and scope questions still go to HITL.`; `eunomia/autonomy/profile.py:20` `class ProjectAutonomyProfile(BaseModel):`.
- **§10 · Brownfield onboarding**: recorded. ADR-0025 keeps the PRD aspirational and stores implementation reality in separate BROWNFIELD_* artifacts, and requires cited evidence_paths to resolve (unresolvable citations are 'rejected or downgraded', not only dropped). eunomia/brownfield/targeted_evidence.py's docstring says it runs a feature-derived search before a feature is declared missing. Excerpts: `docs/v2/adr/0025-brownfield-onboarding.md:26` `PRDComposer remains aspirational: the PRD says what should exist. It`; `docs/v2/adr/0025-brownfield-onboarding.md:82` `that actually resolve in the cloned repo (unresolvable citations are`; `eunomia/brownfield/targeted_evidence.py:5` `specified PRD feature.  Before declaring a feature missing, this module runs a`.
- **§10 · GitHub push webhook with self versus external distinction**: recorded. Found at eunomia/integrations/github_webhook.py; the top-level integrations/ directory holds only figma-plugin and open-design (tree listing). Its docstring classifies push/pull_request deliveries as self (advance the baseline only) or external (re-run the brownfield analyzer), deduplicated by the resulting HEAD SHA (code item 16). The laptop-push use case is the document's own framing. Excerpts: `eunomia/integrations/github_webhook.py:5` `self  → an upstream change Eunomia itself produced (a merged PR that`; `eunomia/integrations/github_webhook.py:9` `external → a human (or another tool) moved the default branch. Refresh`.
- **§10 · Chatbot file generation**: recorded. chatbot/registry.py registers file_gen.pdf, .docx, .pptx, .xlsx and .png (code item 17), with generators in eunomia/chatbot/file_generators/ (tree listing), and a save_generated_file mutation described as saving into an approved project folder (code item 18). Excerpts: `eunomia/chatbot/registry.py:740` `"file_gen.pdf": ToolSpec(`; `eunomia/chatbot/registry.py:880` `"file_gen.png": ToolSpec(`; `eunomia/chatbot/registry.py:422` `"Save a chatbot-generated binary file into an approved project "`.
- **§10 · Steward memory and supervisory lane**: recorded. The cited doc turns approved PRD/tech-spec components into semantic memory and applied decisions into episodic memory, uses project-scoped retrieval with strict result caps, invalidates memories on artifact replacement, and deploys a second worker allow-listed to ProjectSteward. Excerpts: `docs/v2/project-steward-memory-and-supervisory-lane.md:33` `Approved PRD and technical-specification components become semantic memory; applied decisions`; `docs/v2/project-steward-memory-and-supervisory-lane.md:35` `embeddings, with strict result caps. Artifact replacement invalidates memories derived from the old`; `docs/v2/project-steward-memory-and-supervisory-lane.md:23` `EUNOMIA_WORKER_AGENT_ALLOWLIST=ProjectSteward`.
- **§10 · Architectural-layer dispatch phasing**: partly. eunomia/routing/layers.py (AUD-028) sets out ui → service → integration phasing with a drain-and-audit gate, where unclassified tasks never block and the starvation floor overrides the gate (code item 33). It is not the active scheduler policy at this commit: eunomia/registries/routing.yaml ships layer_phasing_enabled: false (code item 31), and the loader defaults to False (code item 32). Excerpts: `eunomia/routing/layers.py:4` `integration work interleaved by score and age alone. Nothing expressed "finish`; `eunomia/routing/layers.py:5` `the interface before wiring the backend to it".`; `eunomia/registries/routing.yaml:30` `layer_phasing_enabled: false`.
- **§10 · Fleet resource snapshots and retained logs**: recorded. ADR-0035 has workers publish CPU, memory, disk and process-memory samples with the 30-second heartbeat, and retains logs in Loki behind an authenticated platform-admin query proxy. eunomia/observability/resources.py (ResourceSnapshot), loki.py and web/src/pages/PlatformLogsPage.tsx exist (tree listing), and resource monitoring has a setting, resource_monitoring_enabled, default true (code item 19). Excerpts: `docs/v2/adr/0035-operational-observability.md:19` `Workers sample CPU, memory, disk, and process memory and publish the latest`; `docs/v2/adr/0035-operational-observability.md:27` `Alloy. Eunomia exposes an authenticated platform-admin query proxy; browsers`; `eunomia/observability/resources.py:13` `class ResourceSnapshot:`.
- **§10 · Non-destructive artifact integrity cron**: recorded. eunomia/workspace/integrity.py's docstring calls it a non-destructive reconciliation of artifact rows against durable content, and its cited line flags content_hash_mismatch; integrity_cron.py schedules it on the master, and config.py enables the sweep by default every 3,600 s (code items 19 and 20). Excerpts: `eunomia/workspace/integrity.py:1` `"""Non-destructive reconciliation of artifact rows against durable content."""`; `eunomia/workspace/integrity.py:54` `reason="content_hash_mismatch",`; `eunomia/workspace/integrity_cron.py:1` `"""Master-side scheduled artifact integrity reconciliation."""`.
- **§10 · In-app documentation the guide reads**: partly. web/src/pages/docsContent.ts contains both sections, 'Investigate before retrying' and 'When the lifecycle looks stuck'. The claim that the guide reads them is not supported: docsContent.ts is imported only by DocsPage.tsx (code item 34), and search_eunomia_help searches the product catalog (code item 35), whose entries link /docs routes rather than hold doc content (code item 36). Excerpts: `web/src/pages/docsContent.ts:406` `heading: "Investigate before retrying",`; `web/src/pages/docsContent.ts:417` `heading: "When the lifecycle looks stuck",`; `eunomia/product_catalog.py:127` `haystack = " ".join((item.route_key, item.title, item.purpose, *item.goals)).lower()`.
- **§10 · Operator mobile app**: partly. The cited path eunomia/mobile/ does not exist; the Expo app is at the top-level mobile/, and eunomia/mobile_builds/ holds the backend for EAS builds (tree listing). Its README describes an Expo/React Native client mirroring the web product's capabilities (code item 37), which is broader than an approvals-and-overview companion. Excerpts: `mobile/README.md:3` `Expo/React Native client for Eunomia. The app mirrors the web product's capabilities using mobile-native list-to-detail workflows and entitlement-aware navigat…`; `mobile/README.md:28` `- Project stack: overview, brownfield, lifecycle, reviews (PRD, technical spec, epic plan, story plan), tasks, bugs, inbox, discussion, workspace, design, arti…`.

#### Section 13.2

- **§13.2 · Eunomia graph practices**: partly. Supported: D13 (only the Sequencer sees the DAG), the cascade walking the supersede chain forward from each affected scope (barrier.py signal_phase), BRD-009..011 typed contracts rendered to Excalidraw with stable element IDs, as-built architecture from repository scans (BRD-016, BRD-021), and BR Auto-mapper embedding rank (BRM-002). Weaker: D16 is Layer-1 hashing only (WS-010 cross-run diffing and WS-013 embedding match unchecked); ADR-0021 lineage is optional per artifact (see 10-L607); F-OVERLAP is defined as 'Scope-glob collision' in CLAUDE.md, but the planner's sibling-overlap check emits ClarificationIssues and the runtime overlap watchdog (SEQ-014) is unchecked. Excerpts: `docs/v2/adr/0013-worker-queue-decoupling.md:12` `**Workers never query the Sequencer.** The Sequencer is the only component that sees the full DAG.`; `FEATURES.md:1004` `- [x] **BRD-010** Deterministic semantic-model → editable Excalidraw renderer with stable element IDs and agent-readable relationship summaries`; `FEATURES.md:1015` `- [x] **BRD-021** Repository-derived brownfield/as-built intelligence scans manifests, routes, migrations, deployment definitions, and completed WorkProducts`.

#### The build list

- **Build list R1 · Clarification queue**: recorded. Only the Eunomia source is checkable here: D17 exists as ADR-0017 and FEATURES.md's CI section. Linear SLAs are outside this repository and were not checked. Excerpts: `docs/v2/adr/0017-clarification-issue.md:1` `# ADR-0017 — ClarificationIssue: the cross-agent uncertainty mechanism`; `FEATURES.md:1057` `### CI — ClarificationIssue lifecycle (D17)`.
- **Build list R2 · Priority inbox for Elikem**: recorded. The Eunomia two-pane Inbox exists as web/src/components/inbox/TwoPaneInbox.tsx, used by InboxPage.tsx and the project InboxTab.tsx (code item 21). Linear Priority Inbox is outside this repository. Excerpts: `web/src/components/inbox/TwoPaneInbox.tsx:17` `export function TwoPaneInbox({`; `web/src/pages/InboxPage.tsx:3` `import { TwoPaneInbox } from "@/components/inbox/TwoPaneInbox";`.
- **Build list R3 · Eve Steward: delegated authority under policy**: recorded. ADR-0040 exists, and PST-001 through PST-010 each appear exactly once in FEATURES.md, all checked. Excerpts: `docs/v2/adr/0040-project-steward-delegated-authority.md:1` `# ADR-0040 — Project Steward delegated authority`; `FEATURES.md:1439` `### PST — Project Steward (ADR-0040)`; `FEATURES.md:1450` `- [x] **PST-010** Project settings controls with regulated-mode restriction and explicit human fallback copy`.
- **Build list R4 · Triage before Elikem**: recorded. The Eunomia BR Auto-mapper is the BRM section of FEATURES.md (BRM-001..007, all seven checked) and eunomia/agents/br_mapper/. Linear triage rules and intelligence are outside this repository. Excerpts: `FEATURES.md:1413` `### BRM — BR Auto-mapper`; `eunomia/agents/br_mapper/README.md:1` `` # `agents/br_mapper/` — BR Auto-mapper ``.
- **Build list R6 · Planning chain from an Eve document**: recorded. PRD, TSC, EPL, STL and ATL are FEATURES.md sections with matching agents under eunomia/agents/ (prd_composer, tech_spec_composer, epic_planner, story_planner, atomic_planner), and the Sequencer has its own SEQ section and eunomia/agents/sequencer/. Excerpts: `FEATURES.md:1086` `### PRD — PRDComposer`; `FEATURES.md:1191` `### ATL — AtomicPlanner`; `FEATURES.md:1218` `### SEQ — Sequencer`.
- **Build list R7 · Change cascade from document revisions**: recorded. D3, D4 and D16 exist as ADR-0003, ADR-0004 and ADR-0016. D16 cross-run identity preservation is only partly built (see 3-L85). Excerpts: `docs/v2/adr/0003-cascade-scope.md:1` `# ADR-0003 — Cascade scope: tagged dependencies, STALE never delete`; `docs/v2/adr/0004-cascade-barrier.md:1` `# ADR-0004 — Cascade barrier: abandon-on-progress, four-phase commit`; `docs/v2/adr/0016-feature-identity-hash.md:1` `# ADR-0016 — Feature identity_hash: stable IDs across PRD regenerations`.
- **Build list R8 · Typed work product on done**: recorded. D15 exists as ADR-0015 with the WP section in FEATURES.md. Linear Guided Reviews are outside this repository. Excerpts: `docs/v2/adr/0015-work-product.md:1` `# ADR-0015 — WorkProduct artifact: structured fields only, one free-form slot`; `FEATURES.md:1266` `### WP — WorkProduct (D15)`.
- **Build list R9 · Certification committee per family and release**: recorded. ADR-0043 exists, and AUD-001 through AUD-040 each appear exactly once in FEATURES.md; all 40 are checked. Excerpts: `docs/v2/adr/0043-certification-requires-audit.md:1` `# ADR-0043: Project certification requires an audit verdict`; `FEATURES.md:1487` `### AUD — Certification audit committee (ADR-0043)`; `FEATURES.md:1534` `- [x] **AUD-040** Hybrid finding identity`.
- **Build list R10 · Family health and agent-drafted updates**: recorded. ADR-0022 exists, and FEATURES.md's PG section implements it. Linear initiatives and project updates are outside this repository. Excerpts: `docs/v2/adr/0022-project-progress-burndown.md:1` `# ADR-0022 — Project progress as burn-down, not percentage`; `FEATURES.md:1467` `### PG — Project progress (burn-down per ADR-0022)`.
- **Build list R12 · Board page: the command centre**: recorded. All three Eunomia surfaces exist: web/src/pages/BoardsPage.tsx, MyQueuePage.tsx and StateOfAffairsPage.tsx. FEATURES.md's BRD section records Eunomia's Boards as Excalidraw brainstorm/architecture boards, not saved Linear-style board views. Linear views are outside this repository. Excerpts: `web/src/pages/BoardsPage.tsx:789` `export default function BoardsPage() {`; `web/src/pages/MyQueuePage.tsx:94` `export default function MyQueuePage() {`; `web/src/pages/StateOfAffairsPage.tsx:47` `export default function StateOfAffairsPage() {`.
- **Build list R14 · Per-lane allowances and fallback chains**: recorded. CLI-007..010 are all checked in the CLI section, and the LLM Control Plane section lists LLM-CP-001..006, all checked. Excerpts: `FEATURES.md:804` `- [x] **CLI-007** Durable usage ledger`; `FEATURES.md:807` `- [x] **CLI-010** Provider execution accounting`; `FEATURES.md:20` `## LLM Control Plane`.
- **Build list R17 · Work-request intake with routing**: recorded. ADR-0038 exists and is implemented by WRQ-001..010. Linear Asks is outside this repository. Excerpts: `docs/v2/adr/0038-post-creation-work-intake.md:1` `# ADR-0038 — Post-creation work enters through workspace artifacts`; `FEATURES.md:1348` `### WRQ — Post-creation work intake (ADR-0038)`.
- **Build list R19 · Hosted review IDE**: recorded. ESP-015 through ESP-019 exist and are checked (code-server IDE, native AtomicTask IDE bridge, EC2 IDE path, remote/headless providers, artifact-only evidence workspaces). Excerpts: `FEATURES.md:1391` `- [x] **ESP-015** code-server human IDE runtime`; `FEATURES.md:1395` `- [x] **ESP-019** Artifact-only evidence workspaces + signed artifact URLs`.
- **Build list R21 · Notes and brainstorm intake with a readiness signal**: recorded. NOT, BRD and RDY are FEATURES.md sections: NOT-001..005 checked with NOT-005b open; BRD-006/007 (optional) open; RDY-003 open. Excerpts: `FEATURES.md:980` `### NOT — Notes (draft → promote) — Phase 2 ingest`; `FEATURES.md:991` `### BRD — Brainstorm boards (Excalidraw) — Phase 2 ingest`; `FEATURES.md:1112` `### RDY — ReadinessAgent`.
- **Build list R25 · Operations projection**: recorded. The Eunomia source exists as eunomia/operations.py, whose docstring calls it the durable application-wide operation projection, with operation_reporter.py and api/routes/operations.py alongside (tree listing); see 10-L591. The 'Where' targets are on the Eve side. Excerpts: `eunomia/operations.py:1` `"""Durable application-wide operation projection and transition helpers."""`.
- **Build list R26 · Knowledge-gap queue from feedback**: partly. The claim text is cut off after 'From Eunomia', so no source is named. By content, the closest Eunomia match is the assistant knowledge-gap records in eunomia/api/routes/assistant_feedback.py: its failure categories include no_knowledge and no_tool, and its triage statuses include documented and product_change (code item 38). Grouping by skill and triage into board tasks have no counterpart there: a gap carries a screen route_key. Excerpts: `eunomia/api/routes/assistant_feedback.py:24` `"no_knowledge", "no_tool", "permission", "ambiguous", "runtime_failure", "unhelpful"`; `eunomia/api/routes/assistant_feedback.py:26` `GapStatus = Literal["open", "investigating", "documented", "product_change", "dismissed"]`.
- **Build list R27 · Advance verb with a refusal matrix**: recorded. The source name is cut off, but the listed refusals (open question, pending review, failed work, no remote, no worker) match the refusal matrix continue_pipeline.py's docstring describes (see 10-L597), and its cited line 223 is the missing_git_remote refusal. Excerpts: `eunomia/lifecycle/continue_pipeline.py:4` `clarifications, rejected/pending gates, failed tasks, missing git remote, no`; `eunomia/lifecycle/continue_pipeline.py:223` `blockers=["missing_git_remote"],`.
- **Build list R28 · Pause a family, drain or immediate**: recorded. ADR-0033's immediate pause releases claimed tasks to ready for retry after resume, and reservations and claims exclude paused projects (queue admission). Excerpts: `docs/v2/adr/0033-project-execution-pause.md:19` `` async boundary and releases claimed tasks to `ready` for retry after resume. ``; `docs/v2/adr/0033-project-execution-pause.md:21` `PostgreSQL is authoritative. Every reservation and direct task claim excludes`.
- **Build list R29 · Pre-publish Git integration**: recorded. ADR-0036 requires the integrated repository to pass the discovered import, build and test gates, then re-checks the base SHA immediately before push. Excerpts: `docs/v2/adr/0036-pre-publish-git-integration.md:26` `3. The integrated repository must pass discovered import, build, and test gates.`; `docs/v2/adr/0036-pre-publish-git-integration.md:27` `4. The base SHA is checked again immediately before push. A moved base retries`.

#### Header and evidence boundary

- **Header · FEATURES.md counts**: differs. At this commit FEATURES.md has 510 checked items (lines matching '- [x]'; no '- [X]') and 32 open items ('- [ ]'), not 506 and 31. The only reconciliation line is the 2026-08-29 note, and that note records '461 checked, 34 outstanding' as of that date; the Status Snapshot table is staler still (Complete 350, Not started 56, Total 414). Excerpts: `FEATURES.md:512` `> **2026-08-29 — ledger reconciled against the code**: the checklist had drifted`; `FEATURES.md:534` `> **461 checked, 34 outstanding.**`; `FEATURES.md:788` `| **Total** | | **414** | **350** | **56** |`.
- **§14 · what the 2026-09-22 read used**: partly. FEATURES.md, CLAUDE.md, README.md and docs/v2/agents-explained.md all exist. The counts differ (510 checked / 32 open at this commit; the 2026-08-29 reconciliation note itself says 461/34). The ADR index (docs/v2/adr/README.md) goes past ADR-0047 to ADR-0048 (QA findings), and the adr/ directory reaches ADR-0049, with a second, unindexed ADR-0048 (Open Design). The index also skips several numbers (0033, 0035, 0036, 0039-0045). Excerpts: `docs/v2/adr/README.md:57` `| [0048](0048-qa-findings-and-retest-granularity.md) | QA findings and per-finding retest granularity | Work pool / Bug reports / Retest / QAP-013…016 | Propos…`; `docs/v2/adr/0049-managed-workspaces-service-modules.md:1` `# ADR-0049: Decompose managed workspaces by AWS service lifecycle`; `FEATURES.md:534` `> **461 checked, 34 outstanding.**`.

#### The cited files at `75f621cf`

The 79 files the excerpts above quote, each with its blob id at
`75f621cf59145167cbfc14055458260cb80a250f`:

- `CLAUDE.md` `93cd5d2bc913bdd0ded364b6862b63b77702ba96`
- `FEATURES.md` `1bb155e0a5d060d61aaf92630f050cfa095d3e99`
- `README.md` `4b46c84d820eeb6deabd6a5603023b3eb05065a4`
- `docs/v2/adr/0003-cascade-scope.md` `050bb79b689c17c0cf231315cae07efc8d601de4`
- `docs/v2/adr/0004-cascade-barrier.md` `492262b5ceca46a5ed6fae076fd294adb632116e`
- `docs/v2/adr/0009-external-input-loop.md` `99b28670ef8c925a564e2f7e8e6497f3c56f28ad`
- `docs/v2/adr/0011-chatbot.md` `d24d9a346af60ce96cd5a2929938640ad51179d3`
- `docs/v2/adr/0013-worker-queue-decoupling.md` `9aea9da82e556c51ffc4583f35ae8ee9863b07bb`
- `docs/v2/adr/0015-work-product.md` `2a906a056348ec7afc534d27e1c05f14bd20b2ec`
- `docs/v2/adr/0016-feature-identity-hash.md` `6764da8a3b55708ebe2a25ddc685fd82a3af906f`
- `docs/v2/adr/0017-clarification-issue.md` `294d99b3b817737ed6d40bdd5bbbcbca0800e082`
- `docs/v2/adr/0021-artifact-lineage.md` `4a7f2aac537e166393c1427359f3d757956eb9c5`
- `docs/v2/adr/0022-project-progress-burndown.md` `c5611435183dc6e54da1505113f28ef4386d3946`
- `docs/v2/adr/0025-brownfield-onboarding.md` `3601c88a042ac47e294ff21c0ef7c0fb5f455eb3`
- `docs/v2/adr/0026-task-review-pr-merge-lifecycle.md` `45a8145fd5e0dacf358994d7f4d2bfe2eeee6d1e`
- `docs/v2/adr/0029-governed-review-proposals.md` `8df63844c3028853030973d9a8fb5ab0ac6e3563`
- `docs/v2/adr/0031-postgres-authoritative-jit-dispatch.md` `b4707362905a098895abfa2266c0384ab4e7dde9`
- `docs/v2/adr/0032-durable-task-stage-executions.md` `847082f01408bbd07866a9771cc9560b22715fce`
- `docs/v2/adr/0033-project-execution-pause.md` `d6ce943dfbf99ee646d91d7a99b4e6af2a2ff69f`
- `docs/v2/adr/0035-operational-observability.md` `3ba4131754e6a463e7ec39e5e1415cc39a936dd7`
- `docs/v2/adr/0036-pre-publish-git-integration.md` `2c5745ab21651794820a8b28f7c09de99b854033`
- `docs/v2/adr/0038-post-creation-work-intake.md` `0f12d72052a40f258782d8ccebc2db80c823acb8`
- `docs/v2/adr/0040-project-steward-delegated-authority.md` `ed9ace874bffd6c899005fbb36a5e5185ba633c4`
- `docs/v2/adr/0042-atomic-planner-finalization.md` `d338572ab29ea30987e60094e1f2fb4c39efa5fb`
- `docs/v2/adr/0043-certification-requires-audit.md` `7e308e5c34268d266513efe54427f94d46862ea4`
- `docs/v2/adr/0045-governed-agent-configuration.md` `98bae66c0ec869d3fd429b319ed6de89c0e27706`
- `docs/v2/adr/0046-design-studio-parity.md` `9e8013009d60aa35c6425b680cba4ae795eab90e`
- `docs/v2/adr/0047-qa-playbook-human-qa.md` `e2a49fdb01fbdf05fc3f675651b70416d460cac6`
- `docs/v2/adr/0049-managed-workspaces-service-modules.md` `f0d48e760371f9fc7d68a8bce0152475b1c9dae7`
- `docs/v2/adr/README.md` `d84f7763fe3113c801ed03e2e77c69d4d534d636`
- `docs/v2/human-resume-jit-dispatch.md` `600f25648d1309bdba203be888ce7f20dd180bb9`
- `docs/v2/operations-inventory.md` `6a4cce37697c9000acf1dcb6eea16e4ec8831acd`
- `docs/v2/project-steward-memory-and-supervisory-lane.md` `5f77641432ff7829044b7c41c28cfea0704766ed`
- `eunomia/agents/atomic_planner/steps/granularity_and_overlap_check.py` `a8d2da85baa044cfa0a466165c901370ccdeccb6`
- `eunomia/agents/br_mapper/README.md` `5fa62c3b97f2d397383fd6bc9e0fdcc4e2954d77`
- `eunomia/api/routes/assistant_feedback.py` `d4179d398ae1d36d09befc9bb6f730b419bb333a`
- `eunomia/api/routes/composer_retry.py` `b1d5cd8ed6df9e340acc19b6f18e619d846eeb7b`
- `eunomia/api/routes/operations.py` `e7e3baddac25ac493ca3aa2bdea0040761e72626`
- `eunomia/api/routes/prd_review.py` `080c886bfef31a2ce0e8a9f89209a2fd5b11c3c0`
- `eunomia/api/routes/recovery.py` `96b6b357c8de4a584932309634fa19e0efa6d905`
- `eunomia/api/routes/state_of_affairs.py` `414fe1641e3f59bc65ad460b7f12256b69995f56`
- `eunomia/autonomy/profile.py` `173270c575546a35cfc582d51a1eb490399ae912`
- `eunomia/brownfield/targeted_evidence.py` `4848f7c5563aa0c502e3a8179b0399be6098e72a`
- `eunomia/chatbot/registry.py` `cc6084303be6dde50b9dc81190dc34b04576c40a`
- `eunomia/clarification/hitl_timeout.py` `76e124d9e28df88b40d304aff0db1dcea512be85`
- `eunomia/clarification/lifecycle.py` `76281a632bc296de81873197710541ce12ee1818`
- `eunomia/core/errors.py` `8fa32eff0d2d7b3c7ae8040151a6a4a94c8e6485`
- `eunomia/core/llm/engines.py` `f7942658bb8e2ecc76a7895987372cf4b47a39cd`
- `eunomia/execution/leases.py` `b58273f6abebb02289bbd3243d576dc49128691f`
- `eunomia/integrations/github_webhook.py` `8361047168749fdae94d0cb0f548e571c57ef56d`
- `eunomia/lifecycle/continue_pipeline.py` `7f2f45c01f8d77b23d0b86ca5e2fb0b95dc058fa`
- `eunomia/llm_control/service.py` `069ba57a3de204bcf9c08972e19ddb2a93a7e31f`
- `eunomia/observability/resources.py` `433d7b26d2bc4a84abc208753728587de40efe2a`
- `eunomia/operations.py` `582bd8af4616c1021f4f971d89911093b63edd09`
- `eunomia/product_catalog.py` `f3a63b051b0fb6734cf1cb25dc6f8a16703bdf9c`
- `eunomia/progress/certificate.py` `ebe6c00036335d5569df2ef96d3210dcfcd00ddc`
- `eunomia/projects/pause.py` `b9aadafd4eb491ed480bc2762e8a658b1043912e`
- `eunomia/registries/lifecycle.yaml` `4c5c9ce2ec388a3032d689a5339167f922425982`
- `eunomia/registries/models.yaml` `1e0ffab0a6fc40d6c7067366cb8e2c89802fb6fa`
- `eunomia/registries/routing.yaml` `225882f3b79307fe48dfe6be4f11ca6cc0222761`
- `eunomia/registries/thresholds.yaml` `cac5844ef1f6c1bd9b0bb27521a0284ee60ba9bd`
- `eunomia/routing/layers.py` `e52cc2f3ae01220d5f1915fbbe9fdaba498261a7`
- `eunomia/schemas/state_of_affairs.py` `e9f3c4c36176c68d953ec446fada0abbd5d472a4`
- `eunomia/workers/claim_loop.py` `60f4f96023d49294eb41c032c808a4a1f61fd564`
- `eunomia/workers/git_integration.py` `bb571285809869d904c799890304651f47994f53`
- `eunomia/workspace/artifacts.py` `dada3d485637237ad154ca035dbcedbc2d09d61a`
- `eunomia/workspace/integrity.py` `b1a10c32bae28aee036bd5079de92f2a9bef63dc`
- `eunomia/workspace/integrity_cron.py` `65562d3d4b5d6e4b95677262071546cb99b31c10`
- `mobile/README.md` `77d4f160636d446113f7164712e7c7877f6cc3c1`
- `web/src/components/guide/EunomiaGuide.tsx` `2f151b3fd1185f02c9494bd910d753117950f115`
- `web/src/components/inbox/SourceMaterialsPanel.tsx` `fd7df428399b2c3b8a7ced64e73ba05f9f0a37fd`
- `web/src/components/inbox/TwoPaneInbox.tsx` `2c6b63fcde4f26920dac4785063cdb268bdac186`
- `web/src/pages/AssistantInsightsSettingsPage.tsx` `daf5bca89a7d5b341c498dafbe2afa93f4d61b15`
- `web/src/pages/BoardsPage.tsx` `aa740a5e5424763ccdad850ba32b51f7bff719a9`
- `web/src/pages/DashboardPage.tsx` `3cb86275c148919335421a78a0be62f11f400774`
- `web/src/pages/InboxPage.tsx` `8c24cfe8ffdb4a120128640dd52b11e5f356b186`
- `web/src/pages/MyQueuePage.tsx` `8a701338fd10345901ad663474c760f9abdfa89b`
- `web/src/pages/StateOfAffairsPage.tsx` `9c29d2aaf5a2734cb70e74cf701e52d2068b5c8b`
- `web/src/pages/docsContent.ts` `b2cc4e944f363c3b4dd7743aa7442419097c08cc`

## 15. Tasks on the Eve board

Created 2026-09-22 through `./eve add` (API mode, `eve-codex-worker-01` credential; the `remote-worker-01` identity was at its rolling 24-hour cap) into family `eve-gap-analysis`, kind `plan`, priority P0, rank 1, task policy `free`.

As the authoring session reported, Elikem's review of the first cut (66 items) found them too coarse for an Opus-class worker. Every epic was therefore decomposed in the same session into atomic sub-tasks: 209 sub-tasks across 60 epics, each naming exact paths, what to do, a done-when clause with the spec or proof to run, its dependencies, and a worker-rules block (touch only the named paths; stop and park rather than invent; no stubs; test-route model binding; evidence with commands and counts). Sub-task ids are `<epic>.<n>`; dependencies run inside an epic and across epics where the report's order requires it. Each epic now depends on all its sub-tasks and its body says it is integration verification only, so `./eve next` offers sub-tasks first and an epic only when every sub-task is verified. The three C items are atomic checks and were left as written; the three L items stay atomic.

Sections on the board: NOW, NEXT, LATER, UI, GRAPH, CHECKS.

Board totals are not recorded here. The family keeps growing as workers and
reviews file sub-tasks, and a count written into this document could not be
checked by its review (review 92e73a5c). The live totals are read with
`./eve board`; the table below maps this document's items to their epics and
sub-tasks, which is what the document relies on.

Board audit of 2026-09-22 (same day, later): every sub-task body was checked against the tree. Twenty-four sub-tasks carried a parent-context block cut at 600 characters and were rewritten with the epic's full text; three named `tools/eve-reviewer/reviewer.mjs`, which does not exist (the reviewer lane is `tools/eve-reviewer/run.mjs`, its report shape `report.mjs`); R34.1 and R21.1 named `libs/oshun/eve-content-core`, which is the package `@oshun/eve-content-core` at `tools/eve-content` (`model.mjs` holds `DOCUMENT_KINDS`); G01.2 and R06.1 named a `buildProductGraphRuntime` export that does not exist (`composeProductGraph` and `readProductGraphArtifact` do); R21.3 and R21.5 cited `publications.ts` and `media.ts` without their `apps/oshun/eve-content/src` path; R01.5 said the board check had eight questions (the shared PostgreSQL board's check has ten: eight in `lib/check.mjs` and a stored-policy and a dependency-target check in `lib/board-check.mjs`); R11.1's parser file was not marked new. All corrected on the board. R33 sits in NEXT (section 12 says NEXT; the first cut filed it under LATER). Tier is now also carried by status: the LATER-tier epics and sub-tasks (R13–R24, R26, R31, R32, L01–L03) and the LATER-tier UI items (U08–U10) are `later`, so `./eve next` offers NOW, NEXT, UI, GRAPH and CHECKS work in position order and the family counted 203 actionable and 72 later, as the authoring session reported on 2026-09-22 (no board report was retained). Every other family on the board was checked in the same pass; the record is the task board polish audit (Eve document `c25ba967-df95-890d-9c38-6d18b21139cd`, revision 1; the file `docs/audits/TASK_BOARD_POLISH_AUDIT_2026-09-22.md` added at `82e9d9a70879`). Rule changes from section 8 that needed no build were applied directly to `AGENTS.md`, `CLAUDE.md` and `.claude/rules/task-checkbox-verification.md`; the wider copy census is epic U11.

| Epic | Section | Sub-tasks (each atomic: paths, do, done-when, deps) |
| --- | --- | --- |
| `eve-gap-analysis:R01` Clarification queue: one typed channel for agent questions | NOW | `R01.1` Questions table and migration on the board (S); `R01.2` Worker API actions ask-question and answer-question (M); `R01.3` CLI: ./eve ask, ./eve answer, ./eve questions, and the brief (S); `R01.4` MCP tools for questions (S); `R01.5` Board check and next: open questions are visible and age (S); `R01.6` Telegram delivery of questions and /answer replies (M); `R01.7` Answered work re-fires with the answer in its brief (S); `R01.8` Human-resume priority in ./eve next (S); `R01.9` Questions as cards in the admin copilot drawer (M) |
| `eve-gap-analysis:R02` Priority inbox for Elikem: one ranked attention queue | NOW | `R02.1` Attention read model composing every source (M); `R02.2` Severity rules and the thresholds file (S); `R02.3` Route GET /v1/workbench/attention (S); `R02.4` Attention page in the Eve Workspace (M); `R02.5` Daily summary at a set hour, delivered and readable aloud (S); `R02.6` Deterministic change summaries for document revisions (S) |
| `eve-gap-analysis:R03` Eve Steward: delegated authority under a per-family policy | NOW | `R03.1` Steward policy schema and per-family storage (S); `R03.2` Append-only steward decision audit table (S); `R03.3` Steward decision engine (shadow first) (M); `R03.4` Delegated verify-task through the steward (S); `R03.5` Delegated document review for non-governed kinds (S); `R03.6` Autonomy profile per family in context and briefs (S); `R03.7` One thresholds file for Eve (S); `R03.8` Shadow week report before any delegation (S) |
| `eve-gap-analysis:R04` Triage before Elikem: rules, suggestions, duplicates, attempt-a-fix | NOW | `R04.1` Intake adapters create draft work items (M); `R04.2` Ordered triage rules applied at intake (S); `R04.3` Task dense index and duplicate/related suggestions (M); `R04.4` Accept, dismiss and per-property auto-apply (S); `R04.5` Family guidance suppresses suggestions (S); `R04.6` Attempt-a-fix lane for crash groups (M) |
| `eve-gap-analysis:R05` Eve Loops: plain-language recurring automations | NOW | `R05.1` Loop definition contract and validator (S); `R05.2` Postgres loop store with versions and run history (S); `R05.3` Loop runner: ticks, event triggers, dedup, execution (M); `R05.4` Write scope enforced on every loop action (S); `R05.5` Loop routes in the BFF (S); `R05.6` Loops page in the Eve Workspace (M); `R05.7` The five first loops as fixtures with a smoke run (S); `R05.8` Explain a run and continue it as a conversation (S) |
| `eve-gap-analysis:R29` Pre-publish Git integration for agent pushes | NOW | `R29.1` Integration tool skeleton: isolated worktree rebase and conflict report (S); `R29.2` Deterministic resolvers for provable shapes (S); `R29.3` Bounded agent resolver for the rest (M); `R29.4` Gates, base-SHA recheck and the two-line push (S); `R29.5` Wire into ./eve done and the automatic worker; document (S) |
| `eve-gap-analysis:R34` Standing guidance document for Eve and per family | NOW | `R34.1` Document kind guidance in Eve content (S); `R34.2` BFF guidance loader with revision pin and cache (S); `R34.3` Inject guidance as the volatile prompt suffix (S); `R34.4` Family-over-global merge with bounds (S); `R34.5` Seed guidance and measure one case (S) |
| `eve-gap-analysis:R06` Planning chain from an Eve document to board tasks | NEXT | `R06.1` Brownfield feature inventory for one root, with evidence rails (M); `R06.2` Planning goal from a PRD document with section anchors (S); `R06.3` Epic and story decomposition with coverage checks (M); `R06.4` Atomic tasks with scope globs, granularity and overlap checks (M); `R06.5` Elikem approves one plan revision; transaction shape (S); `R06.6` Create a family from a brief (S) |
| `eve-gap-analysis:R07` Change cascade from document revisions | NEXT | `R07.1` Revision events feed the cascade (S); `R07.2` SIGNAL and SETTLE: stale marks and paused leases (S); `R07.3` RECONCILE and DISPATCH through the planner (S); `R07.4` Event-driven docs stale replaces the git-blame sweep (S); `R07.5` Remark versus authorised correction on document reviews (S) |
| `eve-gap-analysis:R08` Typed work product on done, with guided review | NEXT | `R08.1` WorkProduct schema and validator (S); `R08.2` ./eve done --product accepts and stores the product (S); `R08.3` Automatic worker composes the product from git and the run (S); `R08.4` Reviewer reads the decision log first; report carries risk and model (S); `R08.5` Guided review view for Elikem (M) |
| `eve-gap-analysis:R09` Certification committee per family and release | NEXT | `R09.1` Committee registry and runner skeleton (S); `R09.2` Floor members from existing scripts (M); `R09.3` Ceiling members: structural and security sweeps on the test route (M); `R09.4` Intent member: feature audit against the family tracker (M); `R09.5` Findings become board tasks with identity dedupe (S); `R09.6` Completion certificate on family closure (S) |
| `eve-gap-analysis:R10` Family health, progress snapshots and agent-drafted updates | NEXT | `R10.1` Family fields: status, health, target date, milestones (S); `R10.2` Append-only progress snapshots (S); `R10.3` Velocity, ETA and lane timing metrics (S); `R10.4` Agent-drafted weekly family update (S); `R10.5` State of Affairs rollup with export (S); `R10.6` Family health shown in the Workspace and the CLI (S) |
| `eve-gap-analysis:R11` Commit and PR automation for board tasks | NEXT | `R11.1` Board-key convention in commits and branches (S); `R11.2` GitHub push and PR webhook receiver (M); `R11.3` State transitions from git events (S); `R11.4` done refuses a commit absent from the remote (S) |
| `eve-gap-analysis:R12` Board views: list, board, timeline with saved filters | NEXT | `R12.1` Saved views model and routes (S); `R12.2` Natural-language filter to a filter tree (S); `R12.3` List and board layouts over the board API (M); `R12.4` Timeline layout (S); `R12.5` Subscribe to a view (S) |
| `eve-gap-analysis:R25` Operations projection: one durable row per long operation | NEXT | `R25.1` Operations table and reporter (S); `R25.2` Wire lanes, loops, watchers and publications to the reporter (S); `R25.3` Acknowledge, retry, cancel and the watchdog (S); `R25.4` Stream and the "what is Eve doing" tool (S) |
| `eve-gap-analysis:R27` `./eve advance <family>` with a refusal matrix | NEXT | `R27.1` ./eve advance <family> with the refusal matrix (S) |
| `eve-gap-analysis:R28` Pause a family: drain or immediate | NEXT | `R28.1` Durable family pause: drain or immediate (S) |
| `eve-gap-analysis:R30` Working context and completion recovery for Eve turns | NEXT | `R30.1` Working-context recorder across turns (S); `R30.2` Completion recovery: bounded second attempt (S); `R30.3` Approval is a checkpoint, not a stop (S) |
| `eve-gap-analysis:R35` Agent session protocol for every lane run | NEXT | `R35.1` Session and activity schema (S); `R35.2` Lanes emit sessions (S); `R35.3` Sessions in the journal, fleet page and metrics (S) |
| `eve-gap-analysis:R36` Releases as objects across code and content | NEXT | `R36.1` Release object and pipeline (S); `R36.2` Release controller records releases (S); `R36.3` Generated release notes and per-pipeline changelog (S) |
| `eve-gap-analysis:R13` Learned skills from journals with a review inbox | LATER | `R13.1` Mining trace selection from journals and reviews (S); `R13.2` Candidate skill extraction with provenance (M); `R13.3` Review inbox: approve, reject, edit (S); `R13.4` Activation feedback loop (S); `R13.5` Deck measurement gate for an admitted skill (S) |
| `eve-gap-analysis:R14` Per-lane allowances, fallback chains and cost reconciliation | LATER | `R14.1` Invocation ledger per lane and purpose (S); `R14.2` Allowances with hard, soft and warn modes (S); `R14.3` Fallback chains with cooldowns and the CLI failure taxonomy (S); `R14.4` Reviewer never the author engine (S); `R14.5` Task-kind turn budgets (S); `R14.6` Billed-stats reconciliation and weekly report (S); `R14.7` Queue monitor and API-boundary rate limiter (S) |
| `eve-gap-analysis:R15` Governed connectors: Eve as MCP client | LATER | `R15.1` Connector registry with encrypted credentials (S); `R15.2` Eve as MCP client: discovery and per-tool review records (S); `R15.3` Read inline, write as card (S); `R15.4` First connectors: GitHub and Hetzner (M); `R15.5` Integration health events (S) |
| `eve-gap-analysis:R16` Cycles for the fleet | LATER | `R16.1` Cycle model and rollover (S); `R16.2` Capacity dial and scope-change graph (S) |
| `eve-gap-analysis:R17` Work-request intake with deterministic routing | LATER | `R17.1` Channel messages become work requests (S); `R17.2` Deterministic routing: direct task, additive plan, new planning run (S); `R17.3` Request form and synced reply thread (S) |
| `eve-gap-analysis:R18` Preview links, browser self-verification and lane spend caps | LATER | `R18.1` Ephemeral preview per PR on the Dev host (M); `R18.2` Browser self-verification loop for lanes (M); `R18.3` Environments per repository root (S); `R18.4` Scoping versus implementation gate and per-lane caps (S) |
| `eve-gap-analysis:R19` Hosted review IDE on the Linux box | LATER | `R19.1` code-server behind Eve auth on the Linux box (M); `R19.2` Artifact-only evidence workspaces (S) |
| `eve-gap-analysis:R21` Notes, brainstorm intake, research drafts and lineage for content | LATER | `R21.1` Notes kind and readiness hint (S); `R21.2` Research drafts with citations, versions and diff (S); `R21.3` Lineage edges for generated artifacts (S); `R21.4` Span attribution and anchored comments (S); `R21.5` Generate document, deck and spreadsheet files as content verbs (S) |
| `eve-gap-analysis:R22` Full-duplex voice (gate: the voice-depth decision record) | LATER | `R22.1` Re-measure the voice-depth decision (S) |
| `eve-gap-analysis:R23` Member request counts on tasks and families | LATER | `R23.1` Aggregate member request counts per task and family (S) |
| `eve-gap-analysis:R24` Task and document templates; duplicate and related relations with merge | LATER | `R24.1` Task and document templates (S); `R24.2` Duplicate and related relations with merge (S) |
| `eve-gap-analysis:R26` Knowledge-gap queue from assistant feedback | LATER | `R26.1` Typed gap records from feedback and abstentions (S); `R26.2` Gap triage into board tasks (S) |
| `eve-gap-analysis:R33` Save a conversation as a skill | NEXT | `R33.1` Save this conversation as a skill (S); `R33.2` Auto-trigger on matching context (S) |
| `eve-gap-analysis:U01` Eve Home: one arrival page for Elikem | UI | `U01.1` Home read model (S); `U01.2` Home page in the Workspace (M); `U01.3` Next move and morning briefing (S) |
| `eve-gap-analysis:U02` One board in the Eve Workspace | UI | `U02.1` Move work-item lanes into the Workspace board (M); `U02.2` Saved filters and family grouping on the board (S) |
| `eve-gap-analysis:U03` Task detail parity with the CLI | UI | `U03.1` path:line, brief and context in the task detail (S); `U03.2` Lease token, expiry, commit link and formatted journal (S); `U03.3` Related documents and structured evidence (S) |
| `eve-gap-analysis:U04` Family pages | UI | `U04.1` Family page with status, health and counts (S); `U04.2` Blocked-by graph and policy editing on the family page (S) |
| `eve-gap-analysis:U05` Fleet page | UI | `U05.1` Fleet page: leases, switches, last runs, stop (M); `U05.2` Spend, allowances and worker resource self-report (S) |
| `eve-gap-analysis:U06` Eve as one conversation everywhere | UI | `U06.1` Full-page Eve chat with history in the Workspace (M); `U06.2` Select-to-context and in-app help (S); `U06.3` Source chips and tool traces; one name on operator surfaces (S) |
| `eve-gap-analysis:U07` One sign-in, one shell for Eve, Operations and the explorer | UI | `U07.1` Session federation design and decision (S); `U07.2` Implement the chosen option (L) |
| `eve-gap-analysis:U08` One visual language with light and dark themes | UI | `U08.1` Workbench-kit tokens in the Eve Workspace with dark mode (S); `U08.2` Light theme for Operations (S) |
| `eve-gap-analysis:U09` Keyboard-first Workspace with unified search | UI | `U09.1` Command palette and list keys in the Workspace (S); `U09.2` Unified search across tasks, docs, decks and chats (S) |
| `eve-gap-analysis:U10` Eve on the phone | UI | `U10.1` Eve tab in admin mobile: attention and questions (M); `U10.2` Confirmation cards and Telegram approvals (S); `U10.3` Task journal and document review on mobile (S) |
| `eve-gap-analysis:U11` Name the human: copy census across Eve surfaces | UI | `U11.1` Census script for human-in-the-loop terms (S); `U11.2` Replace terms in prompts and tool descriptions, re-stamp the hash (S); `U11.3` Replace terms in UI copy with a browser proof (S); `U11.4` Replace terms in Eve ADRs and docs; add the lock (S) |
| `eve-gap-analysis:G01` The Eve graph: one typed property graph in PostgreSQL | GRAPH | `G01.1` Schema eve_graph.node and eve_graph.edge (S); `G01.2` Compile the product graph into eve_graph (S); `G01.3` Compile the board into eve_graph (S); `G01.4` Compile Eve documents into eve_graph (S); `G01.5` Compile work items, decisions and threads (S); `G01.6` Graph version, validation and the check question (S) |
| `eve-gap-analysis:G02` Grounding by neighbourhood everywhere and a graph retrieval arm | GRAPH | `G02.1` Neighbourhood packs for tasks, documents, families, releases (S); `G02.2` Graph retrieval arm and three-arm fusion (S); `G02.3` Measure on the relevance battery (S) |
| `eve-gap-analysis:G03` Impact and cascade as reachability | GRAPH | `G03.1` Forward closure with an edge whitelist (S); `G03.2` Identity hashes for regenerated nodes (S); `G03.3` Snapshot diff (S) |
| `eve-gap-analysis:G04` Scheduling on the DAG: toposort, critical path, overlap | GRAPH | `G04.1` Topological order under priority in next (S); `G04.2` Critical path and blocked-by chains (S); `G04.3` Scope-glob overlap refusal on concurrent leases (S) |
| `eve-gap-analysis:G05` Proposed edges for triage: duplicates, related, caused-by | GRAPH | `G05.1` Proposed edge kinds and the decision route (S); `G05.2` Link prediction boost from shared neighbours (S) |
| `eve-gap-analysis:G06` Write-scope locality for agents, loops and turns | GRAPH | `G06.1` Bounded reachability over eve_graph (S); `G06.2` Enforce locality before a confirmation card (S); `G06.3` Enforce locality in watcher and loop delivery (S) |
| `eve-gap-analysis:G07` Lineage for evidence: derived-from edges with hashes | GRAPH | `G07.1` derived-from edges with hashes for publications and media (S); `G07.2` Backward closure and the certification walk (S) |
| `eve-gap-analysis:G08` The code graph joined to the product graph | GRAPH | `G08.1` Module and import graph compiler from tsconfig paths (M); `G08.2` Join code to product: anchors and visits (S); `G08.3` Blast radius, orphans and hotspots as queries (S) |
| `eve-gap-analysis:G09` Family health as graph metrics | GRAPH | `G09.1` Frontier, chain and stale metrics (S) |
| `eve-gap-analysis:G10` Decision graph constraints | GRAPH | `G10.1` Typed decision outcomes and citation rule (S); `G10.2` Conflict detection (S) |
| `eve-gap-analysis:G11` Shape rules in `./eve check` | GRAPH | `G11.1` Rule runner with fixtures (S); `G11.2` Rules: ready links, superseded citation, shipped commit (S); `G11.3` Rules: verify provenance and proposed-edge age (S); `G11.4` Wire the rules into ./eve check (S) |
| `eve-gap-analysis:G12` Temporal graph queries | GRAPH | `G12.1` Time-sliced queries (S); `G12.2` Since Elikem last looked (S) |
| `eve-gap-analysis:G13` Graph views for Elikem | GRAPH | `G13.1` Blocked-by chain and blast-radius views (S); `G13.2` Lineage view (S) |
| `eve-gap-analysis:L01` Artifact integrity cron for documents and media | LATER | atomic as written |

The epics R20, R31 and R32, the checks C01, C02 and C03 and the later items
L02 and L03 rested only on the excluded Zel material, so their rows are
removed from this table; they remain on the Eve board as filed but are outside
this analysis. The same holds for the parts of retained epics that were filed
from that material: R14.2's soft mode, R14.5, R14.6, R14.7, R15.2's per-tool
review records, R15.5, R21.2, R30.1, R30.3, U01.3's morning briefing, U06.3's
source chips, U09.2 and U10.2's Telegram approvals, and the words "cost
reconciliation", "research drafts" and "unified search" in the titles of R14,
R21 and U09.
