Dashboard

​

When to use

• Visualize the full traceability chain (REQ → UC → WF → API → BDD → INV → ADR → TASK → COMMIT → CODE → TEST)
• Review requirements coverage by use cases, tests, code, and commits
• Identify orphaned artifacts and broken references
• Show stakeholders project health and progress
• Get contextual next-action prompts with one-click copy

​

Invocation

/sdd:dashboard                    # Full scan and HTML generation

​

Fast path with generate.py

python generate.py --project . --output dashboard/

• Large projects (300+ artifacts)
• Repeated dashboard regeneration
• CI/CD integration

• First dashboard generation (to see the process)
• Debugging traceability issues
• The Python script is not available

​

The 5 dashboard views

​

1. Summary view (Executive)

• Health Score hero banner: Letter grade (A-F) with weighted score and 3 actionable recommendations
• Next Action card: Prominent card showing the single most important next step with reason and copy-ready prompt
• Coverage progress bars: Visual progress for REQs with UCs, Code, Tests, and Commits
• Activity Feed panel: Real-time pipeline activity with scrollable feed (updated via JSONP polling every 5s)
• Top gaps: Lists REQs with missing coverage
• Artifact breakdown: Count by type with humanized labels
• Pipeline status: Interactive stage badges with click-to-filter popovers
• Color legend: Explains status dots (Complete, In Progress, Specified, Not Started)

​

2. Matrix view (Traceability)

• Rows: REQs
• Columns: UC, WF, API, BDD, INV, ADR, TASK, CODE, TESTS, COMMITS
• Color-coded cells: green (full), yellow (partial), gray (spec only), red (untraced)
• Click any cell to open detail panel with full chain
• Filters: Domain, Layer, Category, Status

​

3. Classification view (Business Domain)

• Domain cards: Extraction & Processing, CVA Matching, GDPR Compliance, etc.
• Layer breakdown: Backend (FASE 1-6), Frontend (FASE 7-8), Infrastructure (FASE 0)
• Domain stats: REQ count, coverage percentage, functional category breakdown
• Drill-down: Click domain to filter matrix view

​

4. Code Coverage view (Implementation)

• Code stats: Total files, symbols with refs, uncovered symbols
• Test stats: Total test files, tests with refs, framework detected
• Commit stats: Commits with Refs trailers, commits with Task trailers, unique tasks covered
• Symbol table: Lists functions/classes with their artifact refs and call graph data (when code intelligence available)
• Uncovered symbols: Public/exported symbols with no SDD refs
• Process flows: Execution flows mapped to workflows (from GitNexus)

​

5. Adoption view (Onboarding)

• Journey stepper: 4 steps (Onboarding → Reverse Engineering → Reconciliation → Import)
• Scenario card: Detected project scenario with confidence and health score
• Health dimensions: 7 dimensions with per-dimension scores (requirements, specs, tests, architecture, traceability, code quality, pipeline state)
• Findings panel: Critical/high findings from reverse engineering (dead code, tech debt, workarounds)
• Reconciliation panel: Alignment percentage, divergences by type, delta stats
• Import panel: External sources processed, items mapped, quality metrics
• Graceful empty states: Each sub-panel shows “Not available” if the corresponding onboarding skill hasn’t run

​

Interactive features

​

Pipeline stage popovers

• Status and last run date
• Artifact count
• Stage-specific prompt with copy button
• “Filter by Stage” action

​

Contextual prompts

• Next Action card: Full prompt explaining why this action is needed
• Hero recommendations: Each recommendation has its own contextual prompt
• Stage popovers: Stage-specific prompts with current context
• Toast notification: Confirms copy with green toast message

​

Live status polling

• Live dot indicator: Pulsing green (active), yellow (stale >60s), hidden (no file)
• Activity feed updates: New entries appear at top with timestamps
• Stale detection: Shows “Possibly stalled” warning if heartbeat stops while status=running
• Graceful degradation: Silent 404 handling, hidden after 3 failures, race condition guard

​

How it works

​

Process overview

1. Read pipeline state from pipeline-state.json
2. Discover artifact directories (requirements, spec, audits, test, plan, task, src, tests, onboarding, findings, reconciliation, import)
3. Scan onboarding artifacts (5 report types from onboarding skills) → compute adoption stats
4. Extract artifact definitions using patterns from id-patterns-extended.md
5. Extract relationships between artifacts (implements, verifies, orchestrates, etc.)
6. Scan code references in src/ for Refs: comments (JSDoc, inline, decorators)
7. Scan commit references in git history for Refs: and Task: trailers
8. Scan test references in tests/ for artifact refs in test descriptions
9. Classify artifacts by business domain, technical layer, and functional category
10. Build traceability-graph.json with full schema v3 (includes adoption block)
11. Generate HTML files (index.html, guide.html) from templates
12. Generate live status seed file (live-status.js) for real-time feed
13. Open in browser and report metrics

​

Code intelligence enhancement

• Uses codeIntelligence.symbols[] for precise symbol data (no regex heuristic)
• Includes both artifactRefs (direct) and inferredRefs (transitive from call graph)
• Creates inferred-implements relationships with confidence scores
• Enriches codeStats with: symbolsWithInferredRefs, uncoveredSymbols
• Processes GitNexus execution flows and maps to workflows

​

Output artifacts

​

Example output

## Dashboard Generated (v5)

| Metric | Value |
|--------|-------|
| Total Artifacts | 245 |
| Artifact Types | REQ:89, UC:23, WF:12, API:18, BDD:34, INV:15, ADR:8, TASK:46 |
| Total Relationships | 412 |
| REQs with UCs | 87% (78/89) |
| REQs with Code | 62% (55/89) |
| REQs with Tests | 71% (63/89) |
| REQs with Commits | 58% (52/89) |
| Code Files Scanned | 85 (120 symbols with refs) |
| Test Files Scanned | 42 (95 tests with refs) |
| Commits Scanned | 128 (94 with refs, 87 with tasks) |
| Classification | 5 domains, 3 layers |
| Orphaned Artifacts | 3 |
| Broken References | 1 |
| Pipeline Stage | task-implementer (FASE-3) |

Files written:
- `dashboard/traceability-graph.json`
- `dashboard/index.html`
- `dashboard/guide.html`
- `dashboard/live-status.js`

Dashboard opened in default browser.

​

Constraints

• Write only to dashboard/: Never modifies pipeline artifacts
• Tolerate partial pipelines: If only requirements/ exists, still generates a dashboard
• No external dependencies: HTML file is fully self-contained
• Forward-slash paths: Always uses / in file paths, even on Windows
• JSON validity: Output JSON is valid and parseable
• Idempotent: Running multiple times overwrites previous output without side effects

​

Related skills