# Is Oshun (V1–V9) a SOTA Agentic System for High-Quality Content? — A Deep Assessment

**Date:** 2026-06-14 **Scope:** The whole V1–V9 portfolio and the cross-domain
agentic content stack that serves it — game assets, storylines, dialogue, guided
meditations, music, concerts, live commentary, codex/lore, UGC. **Method:**
Direct source reads (no trust of TODO checkboxes or prose claims), five
independent adversarial deep-dive agents (V1 platform; narrative/dialogue; game
assets; wellness/concert/commentary; V8/V9), this session's line-by-line
**strong-verify** of the agentic content-quality stack, and the repo's own
`AGENTIC_CONTENT_QUALITY_ANALYSIS_2026-06-13.md` (its diagnosis) cross-checked
against the now-implemented remediation. Every load-bearing claim carries
`file:line` evidence and an explicit **real / wired / stub / [~]**
classification.

---

## 0. The one-paragraph answer

**The agentic content-_quality engine_ is genuinely state-of-the-art — real,
tested, and rare. The _system as a whole_ is not yet "SOTA content throughout
V1–V9"; it is a SOTA engine with uneven reach.** As recently as 2026-06-13 the
repo's own analysis concluded "**No — not as built today**": every
quality-moving technique (best-of-N, self-refine, reward models, LLM-as-taste-
judge) was either dead code or fabricated its results with
`Math.sin`/`Math.exp`, and a human was the sole arbiter of whether content was
good. The `AGENTIC_CONTENT_QUALITY_TODOS` ledger then **built the real versions
of nearly all of them**, and this session's strong-verify confirmed they are
genuine algorithms with computed-value tests, not stubs. So the _core_ is now
SOTA. But the deep-dives show that core is **wired into some flagship content
paths and not others**: the generic AI Studio service, Hathor narrative, and V4
codex/mission run on it; V6 companion dialogue and Lilith guided meditations are
still template-assembled (and in two places falsely attest "generated"); the V3
concert and V4 commentary quality gates are real but **unconsumed**; asset
generation has real image/3D/audio providers but a **stub** headline 3D-Gaussian
editing library and an **unbuilt** generated-asset→Unreal last mile; and V8/V9
are **specification-only** (0 lines of code). Net: a top-tier quality brain
attached to a partially-connected nervous system.

**Verdict: SOTA core, partial deployment. Confidence: high.**

---

## 1. What V1–V9 actually are

| Ver    | Product (codename)                              | Domain                    | Content it needs                                                                 | Maturity                                                       |
| ------ | ----------------------------------------------- | ------------------------- | -------------------------------------------------------------------------------- | -------------------------------------------------------------- |
| **V1** | Oshun Platform + **Agentic AI Studio**          | Creator/operator platform | Everything — it is the _engine_ + studio that commissions content for all others | **Shipped engine**, real + tested                              |
| **V2** | Fighter game                                    | Combat / Side Story       | Fighter dialogue banks, Side-Story prose, concept art, cooked UE assets          | Game real; content **authored** (not generated)                |
| **V3** | Concert / music experience (Saraswati)          | Live music                | Concert scenes, speeches, music tracks, C2PA-signed exports                      | Music DSP real; quality gate real-but-**unwired**              |
| **V4** | Sports / esports (Calliope)                     | Match play                | Live commentary, codex/lore, mission briefings                                   | Codex/mission **wired** to engine; commentary gate **unwired** |
| **V5** | Open-world RPG                                  | Quests/dialogue at scale  | Side-quests, NPC dialogue (toward ~12k nodes), cinematics                        | Quality-batch path real; volume staged                         |
| **V6** | Cognition / companion AI (Egbe)                 | Living NPCs               | Real-time judged/refined dialogue, chronicle ("Book of Ori")                     | Gateway real; **V6 surfaces are template skeletons**           |
| **V7** | UGC forge (Maya)                                | Creator assist            | Generated assist artifacts behind trust boundary                                 | Forge-assist judge/refine path present                         |
| **V8** | **Ariadne** — self-authoring detective universe | Procedural mysteries      | Fully generated _fair, solvable, voiced_ cases compiled into V5                  | **Spec only** (0/90 tasks, no code)                            |
| **V9** | **Metis** — "a curious ape's guide to reality"  | Consumer learning         | Grounded interactive lessons over real Metis/Nyx/Kalika kernels                  | **Spec only** (0/57 tasks); composed domains real              |

The important structural fact: **V1 is the engine; V2–V9 are consumers.** The
right question is therefore "is the V1 engine SOTA, and does each consumer
actually run its content through it?"

---

## 2. The shared agentic content engine — the SOTA core (real, tested)

This is the part that is genuinely state-of-the-art. It lives mostly in
`libs/shared/content-quality-judge` (32 source files, one per capability),
`libs/shared/content-release-gates`, and `libs/oshun/content-service`, riding on
the real Anthropic provider floor
(`libs/iris/conversation-providers-anthropic/ src/anthropic-provider.ts:307,335`
— real `client.messages.create` + streaming + token counting) exposed through
`@iris/agents-core` with a `BudgetMeter` and `KillSwitchRegistry`.

**The loop.** A first-class
`generate → judge-gate → (regenerate | refine) → re-gate` loop is wired into a
durable, replayable service (`content-pipeline-service.ts:94-164`; dispatcher
`dispatcher.ts:199-322` — generate = best-of-N + self-refine, judge-gate throws
a dimension-targeted critique that the runner feeds back as
`reviewerDirection`). It persists run records + runner checkpoints to disk
(survives restart; Postgres swap behind the same interface), and is mounted
behind a real Node HTTP server + the BFF operator dashboard. Every gated
artifact is stamped with full provenance: run id, model, prompt sha256, content
sha256, per-candidate scores, regenerations, judge score, disagreement
(`content-pipeline-service.ts:135-146`).

**The taste signal (the hard part).** Creative quality has no ground-truth
verifier, so the engine approximates one with a calibrated LLM-judge panel:

| Capability                                                               | Real? / Evidence                                                                                          |
| ------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------- |
| Judge panel (N≥3, median, disagreement→HITL)                             | Real. `judge-panel.ts:71-141` (≥3 enforced; median; spread→`needsHuman`)                                  |
| Position-bias mitigation (pairwise both orderings, consistent-only wins) | Real. `judge-engine.ts:277-341`                                                                           |
| Pointwise mean+variance (≥2 samples)                                     | Real. `judge-engine.ts:96-104` (spec asserts variance 800/3)                                              |
| Calibration vs human gold (Cohen's κ, Pearson, MAE, κ-drift gate)        | Real, textbook-valued tests. `calibration.ts:65-189`; `calibration.spec.ts` asserts κ=0.35135, pe=0.48611 |
| Slop / GPT-ism penalty (EQ-Bench-style, length-controlled, saturating)   | Real. `slop.ts:31-186` (spec asserts weightedHits 6/5/15, penalty=0.5)                                    |

**Inference-time quality.** Best-of-N with **pessimistic** selection
(`LCB = adjustedScore − pessimism·disagreement`, `best-of-n.ts:50-84` — resists
reward-hacking the judge), one-pass structured **self-refine** that keeps a
revision only if it improves and aborts on regression (`self-refine.ts:79-117`),
**Reflexion** lessons carried across batch items via working memory
(`reflexion.ts:88-173`), and per-mode **compute budgets** (fast/balanced/deep/
exhaustive) enforced by a metered provider with kill-switches
(`compute-budget.ts:66-128`).

**Learned taste / continuous improvement.** Real statistics, not bookkeeping:
reward model = the calibrated panel with **reward-hacking detection** via excess
correlation (`reward-model.ts:156-205`; `reward-model.spec.ts` asserts a
length-hacker is flagged), **active learning** with a learned per-bin acceptance
region (`active-learning.ts:89-159`), **drift detection** via direction-aware
Welch z with quarantine recommendation (`drift.ts:71-215`), and
**champion-challenger** promotion on a quality metric via a real two-proportion
z-test (`quality-champion-challenger.ts:55-137`).

**Corpus-level diversity (the level almost nobody touches).** Embedding-space
diversity with mode-collapse detection (`corpus-diversity.ts`), a **batch
release gate** that fails a homogeneous batch even when every item passes
individually (`corpus-gate.ts`), and **diversity pressure** in generation
(novelty score = 1 − max cosine; a MAP-Elites-lite behavior descriptor —
`diversity-pressure.ts`).

**Grounding + model routing + real embeddings.** Real CLIP ViT-B/32 + all-MiniLM
embeddings via `@huggingface/transformers` with **no hash fallback**
(`libs/sophia/embeddings/.../clip-multimodal-backend.ts`,
`text-embedding- model.ts`; 7/7 model-backed tests; the residual SHA-256 image
path was removed this session). A fail-loud **canon grounding gate** that blocks
ungrounded claims, now wired as an opt-in required gate in the content service
(`dispatcher.ts:148-160,257-313`). **Creative model routing** that pins prose/
judging to Opus 4.8 / Fable 5 with auditable fallback (`model-routing.ts`).

**Governance.** Required gates with content-hash binding, evidence-or-fail,
named signoff (`release-gates.ts:124-225`); the adversarial stub scan over the
whole stack returns **zero result-faking hits**.

> This session's strong-verify ran three independent adversarial agents + deep
> reads over all of the above and confirmed every algorithm is real with
> computed-value tests (κ=0.35135, variance 800/3, LCB formula, Welch z, excess
> correlation) — they would _fail_ against a stub. 312+ tests green.

---

## 3. SOTA scorecard: 2026 techniques — _before_ vs _now_

This is the clearest way to see the transformation. The "Before" column is the
repo's own 2026-06-13 audit
(`AGENTIC_CONTENT_QUALITY_ANALYSIS_2026-06-13.md §3`); the "Now" column is the
verified state after the remediation ledger.

| 2026 SOTA technique for creative quality          | Before (2026-06-13)                                | Now (verified)                                        | Evidence                                              |
| ------------------------------------------------- | -------------------------------------------------- | ----------------------------------------------------- | ----------------------------------------------------- |
| Best-of-N (sample, score, pick)                   | **DEAD/FAKED** (`Math.sin` scores)                 | **REAL** (pessimistic LCB)                            | `best-of-n.ts:50-84`                                  |
| Self-Refine / Reflexion                           | **DEAD/TEST-ONLY**                                 | **REAL**                                              | `self-refine.ts`, `reflexion.ts`                      |
| LLM-as-judge of _quality_ (calibrated, de-biased) | **ABSENT** (literary-critic rubrics never called)  | **REAL** (panel + position-bias + slop + calibration) | `judge-engine.ts`, `calibration.ts`                   |
| Reward / preference model + hacking control       | **DEAD/FAKED** (`Math.exp`)                        | **REAL** (panel-as-RM + excess-corr detection)        | `reward-model.ts:156-205`                             |
| Multi-agent narrative collaboration               | **DEAD/TEST-ONLY**                                 | **REAL** (planner/drafter/critic/continuity ensemble) | `narrative-ensemble.ts:89-252`                        |
| Quality-Diversity / MAP-Elites / novelty          | **ABSENT**                                         | **REAL** (corpus diversity + novelty pressure)        | `corpus-diversity.ts`, `diversity-pressure.ts`        |
| RAG grounding bound to generation                 | **PARTIAL** (1 BFF route; Sophia = SHA-256 hashes) | **REAL** (real CLIP/MiniLM; grounding gate wired)     | `clip-multimodal-backend.ts`, `dispatcher.ts:257-313` |
| Slop / length control / debiasing                 | **ABSENT**                                         | **REAL**                                              | `slop.ts`, `external-benchmark.ts`                    |
| Best-creative-model routing                       | **ABSENT** (model id decorative)                   | **REAL** (pinned Opus 4.8 / Fable 5 + fallback)       | `model-routing.ts`                                    |
| Drift / champion-challenger on quality            | **ABSENT**                                         | **REAL** (Welch z; two-proportion z)                  | `drift.ts`, `quality-champion-challenger.ts`          |
| Inference-time thinking budget tied to quality    | **DEAD** (accounting-only)                         | **[~]** (seam exists; live demo gated)                | `claude4-adapter.ts:467` (no live caller)             |
| Fine-tuned/distilled taste model (DPO/RM)         | **DEAD/FAKED**                                     | **[~]** (needs GPU/training infra; fail-loud)         | ledger §5.3                                           |
| The faked `nous/training/*` stack                 | **VIOLATION** (`Math.sin`/`Math.exp` fabrication)  | **GUTTED to fail-loud** `NotConfiguredError`          | ledger Phase 0                                        |

**~10 of ~12 techniques moved from dead/faked/absent → real and tested.** That
is the single most important finding: the engine crossed the line from "governed
mediocrity with a human bottleneck" to a real, calibrated, taste-driven,
self-improving content system. The two remaining are honestly `[~]` (need a live
model demo / GPU), not faked.

---

## 4. Content-domain reality — does "everything" actually run on it?

### 4.1 Storylines, narrative & dialogue — **mostly SOTA; V6 is the hole**

The architecture is right: deterministic planners do _structure_; an LLM writer
surfaces _prose_; judge/lore/continuity/fabrication gates catch failures and
**fail loud**.

- **Canonical narrative (Hathor) — REAL, wired.** `generators.ts:102-308` calls
  a real LLM and throws `NarrativeGenerationError` on empty/partial output; the
  full wire `quality-batch.ts:151-228` runs best-of-N → judge gate →
  lore-consistency → review.
- **Fixed-pool → LLM surfacing — REAL, fail-loud.** `prose-surfacer.ts:130-156`
  throws `FixedPoolLeakError` if the writer parrots a template; the Neith
  `TENSION_VOCAB` banks are correctly used only as the _planner_, never shipped
  (`quest-prose-surfacer.spec.ts:98` asserts the throw).
- **Long-form coherence — REAL.** The multi-agent ensemble beats single-pass
  (`narrative-ensemble.spec.ts` asserts 90 vs 55) and catches a planted
  continuity contradiction (`continuity.spec.ts:36`); the bottom-up simulation
  throws `NarrationFabricationError` if the narrator invents a beat not in the
  sim log (`bottom-up-simulation.spec.ts:110`).
- **V4 codex/mission — REAL, wired** (`v4-narrative-quality.ts`: best-of-N +
  judge gate + self-refine + corpus STOP).
- **V6 Egbe dialogue — TEMPLATE skeleton, NOT wired (the biggest narrative
  hole).** The real cognition gateway exists and is tested (9/9), but V6's
  surfaces never call it: `psyche-agent/src/index.ts:315-348` returns
  `"${agentId} replies from ${groundingState} context: ${promptSummary}"` (no
  LLM); and `cognition-stack/src/index.ts:571-617` returns
  `generatedNatively: true, approved` over **three hardcoded locale strings** —
  a result-faking attestation. `clio-story` ("Book of Ori") is deterministic
  string concatenation labeled with decorative `modelRoute` contracts.
- **V2 Side Story** compiles _authored_ dialogue banks into UE assets by design
  (not generation) — honest, not a stub.

**Verdict: SOTA where wired (Hathor, V4, ensemble); V6 companion dialogue is a
template skeleton that in two places falsely claims generation.**

### 4.2 Game assets (visual / 3D / audio) — **two-tier: real providers, stub headline**

| Asset type                                                            | Status                                                                                                             | Evidence                                                    |
| --------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------ | ----------------------------------------------------------- |
| Concept-art / image gen                                               | **REAL** (FLUX/Stability/RunComfy HTTP + SHA256 integrity + fail-loud)                                             | `bfl-provider.ts:939,958,517`                               |
| 3D mesh gen (text/img→3D)                                             | **REAL** (Meshy/Tripo/Rodin/Hunyuan HTTP, throws on non-200)                                                       | `3d-generation/.../base-provider.ts:183-194`                |
| **3DGS diffusion editing** (gaussctrl/syncnoise/morpheus/intergsedit) | **STUB** — manifest/URI builder, _zero_ numeric computation in 6,143 LOC; tests assert tautologies                 | `syncnoise-…ts:190-203`, `gaussctrl-depth-…ts:220-256`      |
| Avatars (renderer + customization + physics)                          | **REAL** (three.js skinning/blendshapes; Verlet-PBD, computed collision)                                           | `avatar-physics/.../verlet-engine.spec.ts:586` expects 0.51 |
| Music / audio DSP                                                     | **REAL** (RBJ biquad, DX7 FM, Karplus-Strong; Rust 51/51)                                                          | `euterpe/.../audio.spec.ts:854` `linearToDb(0.5)`→−6.0206   |
| USD round-trip / USDZ / cross-DCC                                     | **PARTIAL / shim** — lossy USDA parse; USDZ + cross-DCC are C# codegen; one file hardcodes `true` round-trip flags | `unity-agent/.../usd-round-trip-validation.ts:217-223`      |
| Generated-asset → UE cook/import                                      | **[~] unwired** — real `RunUAT` + UE Python import primitives exist, but **nothing dequeues** the cook job         | `bellona/unreal/.../cook-runner.ts:124-244` (no caller)     |

**Verdict: partially SOTA. Image/3D provider integration, avatars, and audio DSP
are genuinely strong; but the flagship "3DGS diffusion editing" is a
manifest-only stub, USDZ/cross-DCC are codegen shims, and the
generated-asset→Unreal last mile is unbuilt.**

### 4.3 Guided meditations / sleep / daily wellness (Lilith) — **NOT SOTA today**

The named "content services" are delivery/scheduling layers, not generators
(they correctly _fail loud_ with `no_content_available` rather than fabricate —
`daily-content-service.ts:148-159`). The actual script path is **template /
phrase-bank assembly**: `meditation-script-service.ts` has no LLM/fetch refs,
and the production caller POSTs with **`useLLM: false`**
(`autonomous-daily-content-service.ts:365`) — the real LLM draft path exists but
is _disabled by the caller_. Quality is gated by a real _deterministic_ scorer
(Flesch reading-ease), **not** the shared judge panel. The DSP for binaural/
ambient/procedural instruments is **real** and the RAG embedder is **real but
dormant** (no wellness service constructs the pipeline). One orphaned file
(`meditation-generation-pipeline.ts:773,1069`) uses `Math.random()` for "RAG
relevance" — a genuine stub off the production path.

**Verdict: NOT SOTA. Real DSP + a real (disabled) LLM path + real (unwired) RAG;
the meditation/sleep _script text_ is template-assembled and judged only
deterministically.**

### 4.4 Concerts / music (V3 / Saraswati) — **SOTA synthesis, unwired gate, fake signing**

Music synthesis is unequivocally **real DSP** (Rust `neith-synth` 51/51:
wavetable, DX7 FM, subtractive/granular; plus `euterpe/audio-engine`). The
**scene/speech quality gate is real and excellently tested** — every draft
scored by the shared judge panel, corpus diversity + slop, and `promote()`
_throws_ when a scene scores 38 even with valid C2PA
(`concert-scene-quality-gate.spec.ts:186- 209`). **Two gaps:** (1) the gate has
**zero consumers** outside its own tests — it is a real capability not yet
spliced into a running export flow; (2) the concert-track C2PA path is **fake
signing** — `track-c2pa-manifests.ts:216` `stableSha256Hex` is **FNV-1a**
(`0x811c9dc5`/`0x01000193`) mislabeled "sha256", with shape-only verification (a
real Ed25519+SHA-256 signer exists but serves the 3D-asset pipeline, unconnected
here).

**Verdict: SOTA on synthesis and on gate _design_; NOT SOTA end-to-end — the
gate is unwired and track signing is fake.**

### 4.5 Live commentary (V4 / Calliope) — **SOTA in design, not in deployment**

Commentary is genuinely LLM-generated with structure pinned by real domain math
(`moment-scoring.ts`: per-event weights, multi-kill multiplier, logistic
win-probability); the writer **throws on invented plays / skipped moments**
(`commentary-generator.ts:217,194`); a real bias-review requires verbatim
quotes. The quality gate is **real** (deterministic non-repetition always runs;
energy/moment-fit via the judge panel; honestly _unscored_ without one) and
computed-value tested (`commentary-quality-gate.test.ts:136-138`: a
high-judge-but-repetitive package stays **blocked**). **Gap:** the claimed "Rust
broadcast service" consumer **does not exist** (zero `.rs` in `libs/calliope`);
the gate is real but **unconsumed**.

**Verdict: SOTA generation + scoring + gating logic; `[~]` on the production
(Rust service) consumer, which is not present.**

---

## 5. Per-product reach — which products actually run on the engine

| Product                                | Runs content through the SOTA engine?                                               |
| -------------------------------------- | ----------------------------------------------------------------------------------- |
| **V1 Studio**                          | **Yes** — the engine _is_ the product; wired into `content-service` + BFF dashboard |
| **V4 codex/mission**                   | **Yes** — `v4-narrative-quality` (best-of-N + judge + refine + corpus)              |
| **Hathor narrative (V5 quests, etc.)** | **Yes** — `quality-batch` full wire                                                 |
| **V4 commentary**                      | **Gate real, but no production consumer**                                           |
| **V3 concert**                         | **Gate real, but unwired into export; track signing fake**                          |
| **V5 dialogue at volume**              | Path exists (best-of-N + judge + corpus + HITL); volume staged, not yet scaled      |
| **V6 companion dialogue**              | **No** — template skeletons; gateway never called; false "generated" attestations   |
| **V7 forge-assist**                    | Judge/refine path present behind the trust boundary                                 |
| **Lilith meditations**                 | **No** — template scripts; LLM disabled (`useLLM:false`)                            |
| **V8 Ariadne / V9 Metis**              | **Spec only** (0 code); V9 explicitly mandates reusing the shared gate suite        |

The pattern is unmistakable: **the engine is uniformly _available_ but unevenly
_applied_.** Several teams built a real, test-passing quality gate for their
domain and then **did not wire it into the running pipeline** (concert,
commentary), or never connected the real generator at all (V6, meditations).

---

## 6. The level-of-abstraction quality matrix

Quality lives at altitudes; here is where the system now stands at each (the
engine raised most of these from "nobody" to "real"):

| Level                     | "Good" means                    | Status                                                                                                 |
| ------------------------- | ------------------------------- | ------------------------------------------------------------------------------------------------------ |
| Token / phrase            | no slop, fresh diction          | **Real** (slop list + maintenance job)                                                                 |
| Sentence / voice          | natural, distinctive _and good_ | **Real** (judge naturalness dim + voice-quality dim)                                                   |
| Passage / prose           | imagery, pacing, beats land     | **Real** (judge panel + self-refine)                                                                   |
| Artifact (quest/scene)    | coherent, on-spec, canon-safe   | **Real** (validators + grounding gate)                                                                 |
| Arc (10k+ words / season) | long-range coherence, payoff    | **Real** (multi-agent ensemble + arc-coherence judge)                                                  |
| Corpus (whole catalog)    | diversity, not 12k clones       | **Real** (corpus diversity gate + novelty pressure)                                                    |
| Player-experienced        | fun / moving in play            | **Real proxy** (player-proxy evals: a tedious quest scores lower) — pre-ship proxy, not live telemetry |

This is the rarest part. Most "AI content factories" have _nothing_ at the
corpus and player-experienced levels; Oshun has real, tested machinery at both.

---

## 7. Strengths — what is genuinely SOTA and uncommon

1. **A real, calibrated taste signal** for a domain with no ground-truth
   verifier — judge panel with position-bias mitigation, slop control, and
   κ-calibration against human gold. Most teams ship an uncalibrated single
   judge.
2. **Pessimistic best-of-N** that explicitly resists reward-hacking — the exact
   mitigation the 2026 LitBench literature calls for, implemented and tested.
3. **Corpus-level diversity as a release gate** — fails a homogeneous batch even
   when every item passes individually. Almost nobody addresses mode collapse at
   catalog scale.
4. **Honesty discipline as infrastructure.** A fail-loud everywhere posture
   (`NotConfiguredError`, `FixedPoolLeakError`, `NarrationFabricationError`), a
   pre-commit stub-indicator scan, and the gutting of the previously-faked
   `nous/training/*` stack. The system prefers to _refuse_ over _fabricate_.
5. **Full provenance + governance** (run id, model, prompt/content sha256, judge
   scores, budgets, kill-switches, named signoff, content-hash binding) — the
   safety/auditability layer is production-grade.
6. **Plan-then-write + mandatory grounding** is already aligned with 2026
   game-PCG SOTA; prose can't invent canon.

---

## 8. Gaps & risks — the honest ledger

**The wiring gap (the dominant theme).** The single biggest distance between
"SOTA engine" and "SOTA system" is that real quality gates are **built but not
consumed**:

- Concert scene-quality/diversity gate: no production consumer.
- Commentary quality gate: no Rust (or any) consumer.
- Grounding gate: real and (this session) wired into `content-service`, but not
  into product-specific flows.

**Real generators left disconnected.**

- **Lilith meditations** call the generator with `useLLM: false`; the real LLM
  and the real RAG embedder are present but disconnected.
- **V6 companion dialogue** never calls the cognition gateway.

**Honesty defects found outside the strong-verified ledger** (these are _new_
findings of this analysis and were not in the agentic-quality ledger's scope —
they should be fixed):

- `cognition-stack/.../index.ts:571-617` returns `generatedNatively: true` over
  a 3-locale hardcoded template — a result-faking attestation.
- `track-c2pa-manifests.ts:216` mislabels FNV-1a as "sha256" and does shape-only
  verification — fake content signing.
- `usd-round-trip-validation.ts:217-223` hardcodes
  `exportsSceneToUsd/ reimportsUsdStage = true` regardless of input.
- `meditation-generation-pipeline.ts:773,1069` uses `Math.random()` for RAG
  relevance (orphaned, off the production path).

**Stub / shim assets.**

- 3DGS diffusion editing (`libs/isis/3dgs-diffusion-editing`) is manifest-only —
  no depth rasterization, ControlNet inference, or shared-noise prediction.
- USDZ / cross-DCC USD are C# codegen shims with no transform math.

**External-gated `[~]` (honest, not faked).**

- Generated-asset → Unreal cook/import last mile (real primitives, no
  orchestrator).
- Trained taste model (DPO/RM) — needs GPU/training infra.
- Live extended-thinking-budget demonstration — needs a live model.
- On-engine cooks, certification, VO/voice, art at volume.

**Roadmap.** V8 (Ariadne) and V9 (Metis) are **specification-only** (0/90 and
0/57 tasks, no code). V9 is the more de-risked (its composed domains — Metis,
Nyx, Kalika, Mnemosyne, Sophia — are real) and explicitly mandates reusing the
shared gate suite; V8's generative core (the case generator) is confirmed
**missing**.

---

## 9. What it would take to make the full claim true

In rough priority order (each is "wire the real thing that already exists," not
"invent"):

1. **Consume the gates you already built.** Splice the concert quality/diversity
   gate into the V3 export flow; put the commentary quality gate in front of the
   broadcast path. (Pattern proven this session by wiring the grounding gate
   into `content-service`.)
2. **Connect the real generators.** Flip Lilith to `useLLM: true` behind the
   judge panel + the (already real) RAG embedder; route V6 companion dialogue
   through the cognition gateway instead of template strings.
3. **Fix the four honesty defects** (V6 `generatedNatively`, FNV-1a "sha256",
   hardcoded USD round-trip flags, `Math.random` RAG relevance) — either make
   them real or fail loud.
4. **Replace the 3DGS stub** with a real (or honestly fail-loud) editor; build
   the generated-asset → Unreal cook orchestrator (`[~]` until then).
5. **Run a live calibration pass** (κ vs human gold with real provider creds) so
   the taste signal's quality is _measured_, not just architecturally sound.
6. **Stand up V8/V9 Phase 0** on the shared stack (V9 first — its foundations
   are real).

---

## 10. Final verdict

**Is Oshun a SOTA agentic system for creating high-quality content across
V1–V9?**

- **The quality engine: yes — genuinely state-of-the-art and rare.** As of
  2026-06-13 it was not (the techniques were faked or dead); the remediation
  ledger built real, tested versions of ~10 of the ~12 SOTA techniques for
  creative quality, verified this session against computed values. A calibrated
  taste panel, pessimistic best-of-N, structured self-refine, multi-agent
  long-form, corpus-diversity gating, learned-taste feedback, grounding, and
  best-model routing — all real, all behind fail-loud governance — is a top-tier
  agentic content brain.

- **The system as a whole: not yet — SOTA core, partial reach.** The engine is
  wired into the generic AI Studio service, Hathor narrative, and V4 codex/
  mission; it is **not** wired into V6 companion dialogue or Lilith meditations
  (still template), and the V3 concert and V4 commentary quality gates are real
  but **unconsumed**. Asset generation has real image/3D/audio providers and
  avatars but a **stub** headline 3DGS library and an **unbuilt** asset→engine
  last mile. V8/V9 are specs. There are four discrete honesty defects to fix
  outside the verified core.

**Bottom line:** Oshun has built the hard thing — a real, calibrated, self-
improving _taste_ engine, which is where almost every "AI content factory" fails
— and most of the remaining distance to "high-quality content throughout V1–V9"
is **integration and honest cleanup, not invention**: consume the gates already
written, connect the generators already built, fix four fabrications, replace
one stub, and build one asset→engine bridge. **Confidence: high** (grounded in
direct source reads, five adversarial deep-dives, this session's line-by-line
strong-verify, and the repo's own before/after analysis).

---

### Appendix — confidence & method notes

- "Real" = a domain-specific algorithm with a test asserting a **computed**
  value against a known-correct answer (would fail on a stub). "Wired" = invoked
  from a non-test production path with the relevant parameter reaching a real
  model/DSP. "[~]" = honestly external/live-operation gated. "Stub" = code that
  fabricates a result it did not compute.
- The agentic-quality core (§2, §3) is **high confidence** — line-by-line
  strong-verified this session, 312+ tests run green, adversarial scan clean.
- The domain breadth (§4) is **high confidence** — five independent adversarial
  agents, each source-reading primary files and running the relevant suites.
- V8/V9 (§1, §8) are **high confidence spec-only** — filesystem + git confirm no
  implementing code exists.