Eve library

Yaa Test Coverage Analysis

Guide

Revision 1

On this page

Analysis Date: January 2026 Scope: Complete Yaa codebase test coverage assessment Methodology: File count analysis, test execution, gap identification


Table of Contents#

  1. Executive Summary
  2. Overall Coverage Statistics
  3. Package-Level Analysis
  4. Critical Paths Without Tests
  5. Test Quality Assessment
  6. Prioritized Test Additions
  7. Implementation Recommendations

Executive Summary#

The Yaa codebase has 217 test files covering 1,101 source files, representing a ~20% file coverage ratio. Test execution shows a healthy pass rate with all 55 tests passing in the AI agents package.

Key Findings:

Metric Value Assessment
Total Test Files 217 Moderate
Total Source Files 1,101 Large codebase
File Coverage Ratio ~20% Needs improvement
Packages with 0 Tests 17 Critical gap
Test Pass Rate 100% (sampled) Healthy

Critical Gaps:

  • Ghana NLP package - Core localization with 0 tests
  • TikTok packages - 7 packages with 0 tests
  • Instagram packages - 4 packages with 0 tests
  • Blindspot detection - Critical AI feature with 0 tests
  • Original content generation - Core journalism feature with 0 tests

Overall Coverage Statistics#

Summary by Domain#

Domain Test Files Source Files Coverage Status
packages/ai 58 249 23% Fair
packages/web 42 178 24% Fair
packages/video 28 184 15% Poor
packages/nlp 27 118 23% Fair
packages/ingestion 24 129 19% Poor
packages/social 19 115 17% Poor
packages/api 11 80 14% Poor
packages/audio 6 34 18% Poor
packages/models 2 14 14% Poor
TOTAL 217 1,101 ~20% Fair

Test File Distribution#

text
60+ tests:  0 domains
40-60:      2 domains (ai, web)
20-40:      3 domains (video, nlp, ingestion)
10-20:      1 domain (social)
<10:        3 domains (api, audio, models)

Package-Level Analysis#

AI Packages (58 tests / 249 sources = 23%)#

Package Tests Sources Ratio Priority
agents 31 84 37% Good
story-clustering-worker 3 13 23% Fair
market-summary-generator 3 9 33% Good
weather-report-generator 2 8 25% Fair
political-bias-trainer 2 8 25% Fair
traffic-update-generator 2 9 22% Fair
gpl-score-generator 2 8 25% Fair
article-tagging-worker 2 8 25% Fair
headline-variant-generator 1 8 13% Poor
trend-analysis-generator 1 8 13% Poor
press-release-summarizer 1 7 14% Poor
fuel-price-update-generator 1 9 11% Poor
event-calendar-updater 1 8 13% Poor
ecg-load-shedding-tracker 1 9 11% Poor
claim-linking-worker 1 6 17% Poor
breaking-news-draft-generator 1 8 13% Poor
article-summarization-worker 1 8 13% Poor
article-priority-worker 1 6 17% Poor
source-accuracy-worker 1 5 20% Fair
blindspot-detection 0 8 0% Critical
fact-check-integrations 0 6 0% Critical
original-content 0 6 0% Critical

NLP Packages (27 tests / 118 sources = 23%)#

Package Tests Sources Ratio Priority
claim-extraction-service 3 10 30% Good
embedding-service 3 9 33% Good
evidence-retrieval-service 3 10 30% Good
language-service 3 11 27% Fair
sentiment-service 3 9 33% Good
abstractive-summarization-service 2 9 22% Fair
extractive-summarization-service 2 8 25% Fair
keyword-extraction-service 2 9 22% Fair
similarity-service 2 8 25% Fair
topic-classification-service 2 8 25% Fair
khaya-client 1 4 25% Fair
service-kit 1 6 17% Poor
ghana-nlp 0 17 0% Critical

Video Packages (28 tests / 184 sources = 15%)#

Package Tests Sources Ratio Priority
youtube-shorts-scheduler 4 9 44% Good
article-to-video-service 3 13 23% Fair
video-captioning-service 3 10 30% Good
video-thumbnail-generator 3 10 30% Good
heygen-service 2 9 22% Fair
video-bulletin-scheduler 2 10 20% Fair
video-generation-queue 2 11 18% Poor
video-hls-service 2 11 18% Poor
youtube-shorts-upload-service 2 11 18% Poor
youtube-upload-service 2 15 13% Poor
avatar-profiles 1 7 14% Poor
video-rendering-service 1 8 13% Poor
video-vertical-converter 1 7 14% Poor
instagram-content-templates 0 5 0% High
instagram-reels-producer 0 8 0% High
instagram-stories-producer 0 8 0% High
tiktok-analytics 0 7 0% High
tiktok-scheduler 0 9 0% High
tiktok-sound-trends 0 8 0% High
tiktok-video-producer 0 8 0% High

Social Packages (19 tests / 115 sources = 17%)#

Package Tests Sources Ratio Priority
whatsapp-business-service 5 13 38% Good
tiktok-business-setup 4 9 44% Good
whatsapp-channel-poster 2 6 33% Good
facebook-page-setup 1 7 14% Poor
instagram-business-setup 1 9 11% Poor
linkedin-company-page-setup 1 6 17% Poor
telegram-channel-setup 1 6 17% Poor
tiktok-strategy 1 5 20% Fair
x-account-setup 1 7 14% Poor
youtube-branding-assets 1 5 20% Fair
youtube-channel-setup 1 7 14% Poor
analytics-dashboard 0 6 0% Medium
buffer-integration 0 5 0% Medium
content-calendar 0 7 0% Medium
instagram-scheduler 0 6 0% High
posting-optimizer 0 5 0% Medium
unified-scheduler 0 6 0% Medium

Ingestion Packages (24 tests / 129 sources = 19%)#

Package Tests Sources Ratio Priority
source-discovery 5 19 26% Fair
rss-collector 3 10 30% Good
scraper 3 9 33% Good
media-extraction 2 9 22% Fair
social-x 2 8 25% Fair
content-normalization 1 7 14% Poor
language-detection 1 8 13% Poor
monitoring-dashboard 1 11 9% Poor
processing-queue 1 12 8% Poor
scheduler 1 5 20% Fair
semantic-dedup 1 7 14% Poor
sitemap-collector 1 9 11% Poor
source-attribution 1 6 17% Poor
timestamp-normalization 1 9 11% Poor

Audio Packages (6 tests / 34 sources = 18%)#

Package Tests Sources Ratio Priority
article-to-audio-service 2 10 20% Fair
elevenlabs-service 2 8 25% Fair
audio-generation-queue 1 10 10% Poor
voice-profiles 1 6 17% Poor

API Package (11 tests / 80 sources = 14%)#

Test File Focus Area Status
server.test.ts Basic server startup Present
searchIndexer.test.ts Search indexing Present
searchQueryParser.test.ts Query parsing Present
searchAnalytics.test.ts Analytics Present
searchSuggestionsRoute.test.ts Suggestions Present
searchRouteAdvancedQuery.test.ts Advanced queries Present
searchPersonalizedRanking.test.ts Personalization Present
sourceReliabilityScoring.test.ts Source scoring Present
verificationScoring.test.ts Verification Present
confidenceClassification.test.ts Classification Present
writingAssistantRoute.test.ts Writing assistant Present

Missing Tests:

  • Authentication routes
  • Article CRUD routes
  • Story management routes
  • Claim management routes
  • User management routes
  • Webhook handlers
  • Rate limiting
  • Error handling middleware

Web Package (42 tests / 178 sources = 24%)#

Well-tested server-side logic:

  • Article AI decision making
  • Source preferences
  • Publication scheduling
  • Topic suggestions
  • Feed queries and blending
  • User data management
  • Push notifications
  • OAuth authentication

Missing Tests:

  • React components (frontend)
  • Page routes
  • API route handlers
  • Middleware
  • Client-side utilities

Models Package (2 tests / 14 sources = 14%)#

Test File Focus Area
ghana-political-spectrum.test.ts Political bias scoring

Missing Tests:

  • Database schema validations
  • Type guard functions
  • Enum validation
  • Entity relationship tests

Critical Paths Without Tests#

Tier 1: Business-Critical (Must Have)#

Path Package Risk Impact
Blindspot detection logic ai/blindspot-detection HIGH Core differentiation feature
Fact-check integrations ai/fact-check-integrations HIGH Trust and accuracy
Original content generation ai/original-content HIGH Content production
Ghana NLP TTS/translation nlp/ghana-nlp HIGH Local language support
API authentication api/src/auth HIGH Security
Article publishing flow web/lib/server HIGH Core business logic

Tier 2: Revenue-Critical (Should Have)#

Path Package Risk Impact
TikTok video production video/tiktok-video-producer MEDIUM Social reach
Instagram content production video/instagram-* MEDIUM Social reach
Content scheduling social/unified-scheduler MEDIUM Publishing automation
Analytics dashboard social/analytics-dashboard MEDIUM Metrics visibility

Tier 3: Operational (Nice to Have)#

Path Package Risk Impact
Buffer integration social/buffer-integration LOW Third-party integration
Content calendar social/content-calendar LOW Planning tool
Posting optimizer social/posting-optimizer LOW Optimization feature

Test Quality Assessment#

Positive Indicators#

  1. High Pass Rate - All sampled tests pass (55/55 in ai-agents)
  2. Integration Tests - Agent tests cover multi-component flows
  3. Async Coverage - Tests properly handle async operations
  4. Timeout Handling - Tests have appropriate timeouts

Quality Concerns#

Type Safety Issues (from code quality analysis)#

typescript
// 40+ instances like this in tests:
const data = result.data as any;
expect((created?.data as any).ok).toBe(true);

Impact: Tests may pass but miss type errors

Mock Coverage Gaps#

Many tests mock at high levels without testing:

  • Error handling paths
  • Edge cases
  • Timeout scenarios
  • Network failure scenarios

Missing Test Categories#

  1. Unit Tests - Most packages lack pure unit tests
  2. Contract Tests - No API contract tests
  3. Performance Tests - No load/stress tests
  4. E2E Tests - Limited end-to-end coverage
  5. Snapshot Tests - No component snapshots

Prioritized Test Additions#

P0: Security & Trust (Immediate)#

Package Test Focus Effort Impact
api/src/auth JWT validation, session management M Critical
ai/fact-check-integrations Third-party API integration M Critical
web/lib/server/cronAuth Timing attack prevention S Critical

P1: Core Business Logic (This Sprint)#

Package Test Focus Effort Impact
ai/blindspot-detection Algorithm correctness M High
ai/original-content Content generation quality L High
nlp/ghana-nlp TTS, translation accuracy L High
api routes CRUD operations, validation L High

P2: Social Platform Integration (This Month)#

Package Test Focus Effort Impact
video/tiktok-video-producer Video rendering pipeline M Medium
video/instagram-reels-producer Reel generation M Medium
social/instagram-scheduler Scheduling logic S Medium
social/unified-scheduler Multi-platform coordination M Medium

P3: Quality Improvement (This Quarter)#

Package Test Focus Effort Impact
All packages Remove as any from tests M Medium
api Contract tests with OpenAPI L Medium
web Component tests L Medium
All packages Error handling tests M Medium

Implementation Recommendations#

1. Test Infrastructure Setup#

typescript
// packages/test-utils/src/index.ts
export * from './mocks/database';
export * from './mocks/redis';
export * from './mocks/llm';
export * from './mocks/external-apis';
export * from './fixtures/articles';
export * from './fixtures/users';
export * from './assertions/custom';

2. Coverage Thresholds#

Add to vitest.config.ts:

typescript
export default defineConfig({
  test: {
    coverage: {
      provider: 'v8',
      reporter: ['text', 'json', 'html'],
      thresholds: {
        lines: 60,
        functions: 60,
        branches: 50,
        statements: 60,
      },
      exclude: ['node_modules/', 'dist/', '**/*.test.ts', '**/*.d.ts'],
    },
  },
});

3. Test File Template#

typescript
// {package}/test/{module}.test.ts
import { describe, it, expect, beforeEach, afterEach, vi } from 'vitest';
import { createTestPool, createMockRedis } from '@yaa/test-utils';

describe('ModuleName', () => {
  let pool: Pool;
  let redis: MockRedis;

  beforeEach(async () => {
    pool = createTestPool();
    redis = createMockRedis();
  });

  afterEach(async () => {
    await pool.end();
    redis.disconnect();
    vi.resetAllMocks();
  });

  describe('functionName', () => {
    it('should handle normal case', async () => {
      // Arrange
      const input = {
        /* ... */
      };

      // Act
      const result = await functionName(input);

      // Assert
      expect(result).toEqual({
        /* ... */
      });
    });

    it('should handle error case', async () => {
      // Arrange
      vi.spyOn(pool, 'query').mockRejectedValueOnce(new Error('DB error'));

      // Act & Assert
      await expect(functionName({})).rejects.toThrow('DB error');
    });

    it('should handle edge case', async () => {
      // Test empty input, null values, boundary conditions
    });
  });
});

4. CI Integration#

yaml
# .github/workflows/test.yml
name: Test
on: [push, pull_request]

jobs:
  test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: pnpm/action-setup@v2
      - uses: actions/setup-node@v4
        with:
          node-version: '22'
          cache: 'pnpm'

      - run: pnpm install
      - run: pnpm test -- --coverage

      - name: Upload coverage
        uses: codecov/codecov-action@v4
        with:
          files: ./coverage/coverage-final.json

5. Test Addition Timeline#

Week 1-2: Security Tests

  • api/src/auth tests
  • fact-check-integrations tests
  • cronAuth tests

Week 3-4: Core Business Tests

  • blindspot-detection tests
  • original-content tests
  • ghana-nlp tests

Month 2: Social Platform Tests

  • TikTok packages
  • Instagram packages
  • Scheduling packages

Month 3: Quality Improvement

  • Remove as any casts
  • Add contract tests
  • Add component tests

Appendix: Test Execution Results#

AI Agents Package (Sample)#

text
Test Files  31 passed (31)
     Tests  55 passed (55)
  Start at  14:57:03
  Duration  12.71s

Test Breakdown:
- Editor-in-chief tests: 12 tests
- Managing editor tests: 8 tests
- Journalist tests: 6 tests
- DevOps/QA tests: 4 tests
- Infrastructure tests: 5 tests
- Utility tests: 20 tests

Packages with Zero Tests (17 total)#

AI Domain (3):

  • blindspot-detection
  • fact-check-integrations
  • original-content

NLP Domain (1):

  • ghana-nlp

Video Domain (7):

  • instagram-content-templates
  • instagram-reels-producer
  • instagram-stories-producer
  • tiktok-analytics
  • tiktok-scheduler
  • tiktok-sound-trends
  • tiktok-video-producer

Social Domain (6):

  • analytics-dashboard
  • buffer-integration
  • content-calendar
  • instagram-scheduler
  • posting-optimizer
  • unified-scheduler

Summary Metrics#

Metric Current Target Gap
Test Files 217 400+ 183
File Coverage 20% 60% 40%
Packages with 0 Tests 17 0 17
as any in Tests 40+ 0 40+
Integration Tests Yes Yes -
E2E Tests No Yes Missing
Contract Tests No Yes Missing
Performance Tests No Yes Missing

Analysis Date: January 2026

Part of the Veritas domain integration documentation