feat(#363): add /map-architecture skill for architecture-deepening reports#373
Conversation
…ports Adds the opt-in architecture-deepening report workflow as a new `/map-architecture` skill. The skill ranks codebase areas by recent git hotspot frequency and design-friction signals (shallow interface, low locality, seam leakage, hard-to-test, ADR conflict), writes a ranked Markdown+Mermaid candidate report under `.map/<branch>/architecture-report/`, and waits for the user to pick one candidate before any code changes begin. Key design choices: - Three phases: scope → report → select (no implementation in skill) - Explicit "do not implement yet" guardrail at every decision point - Companion report.json for future eval assertions - Hands off to /map-plan (score ≥ 6) or /map-fast (score < 6) after pick - Deferral notes in .map/architecture-notes.md to avoid re-surfacing Also registers the skill in skill-rules.json with 7 keywords and 3 intent patterns, adds it to WORKFLOW_EFFORT_PROFILES in test_skills.py (effort: medium/adaptive), and updates the skill count guard in test_skills_consistency.py from 19 → 20. Co-Authored-By: Claude <noreply@anthropic.com>
|
Warning Review limit reached
Next review available in: 25 seconds Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (9)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
2 participants
Summary
/map-architectureas a new opt-in skill that ranks codebase areas by recent git hotspot frequency and design-friction signals, generates a Markdown+Mermaid candidate report, and waits for the user to pick one candidate before any code changesmap-architectureinskill-rules.jsonwith 7 keywords and 3 intent patternsWORKFLOW_EFFORT_PROFILESintest_skills.py(medium/adaptive)test_skills_consistency.pyfrom 19 → 20What the skill does
Phase 1 — Scope: accepts an optional
<module-path-or-pain-point>argument; falls back togit log --since="90 days ago"hotspot analysis, filtering out lock files, migrations, and generated output trees.Phase 2 — Report: scores each candidate area on 6 design-friction signals (change frequency, shallow interface, low locality, seam leakage, hard-to-test, ADR conflict; 0–2 each). Writes:
.map/<branch>/architecture-report/report.md— Markdown with Mermaid before/after diagrams per candidate.map/<branch>/architecture-report/report.json— machine-readable companion (candidate count, scores, top recommendation) for future evalsPhase 3 — Select: presents candidates, asks user to pick ONE, then recommends
/map-plan(score ≥ 6 or multi-file) or/map-fast(score < 6, bounded). Optionally records deferral notes in.map/architecture-notes.md.Design guardrails baked into the skill
Test plan
tests/test_skills.py— all 289 skill structural/frontmatter/trigger tests pass (including newmap-architectureentry inWORKFLOW_EFFORT_PROFILES)tests/test_skills_consistency.py— skill count guard updated 19 → 20; all 43 tests passtests/test_template_render.py— byte-identity golden render tests pass (rendered trees match source)make check-render— generated trees matchtemplates_srcCloses #363 (slice 1 — skill workflow instructions + registration; Python tooling for report generation is optional future work per issue guardrails)
Generated by Claude Code