claude/skills/ship-feature/SKILL.md
bastien 66beae16b7 auto-detect plan, complexity scoring, ctx7 cache, graphify in workflows
- detect_plan() auto-detects Max/Pro/Free from ~/.claude.json
- session-start budget adapts to plan (Max=20k, Pro=11k, Free=5k)
- token counting now uses only ACTIVE plugins, not installed binaries
- statusline shows plan label + session duration instead of start time
- plugin-advisor: complexity assessment (0-100%) drives tool selection
- plugin-advisor: auto-activation with confirmation (PHASE 4)
- ruflo OFF by default, GSD v2 preferred for multi-session
- init-project: ctx7 pre-fetch + graphify scaffold + graphify full
- ship-feature: ctx7 cache check before implementation
- frontend-design disabled in installer (doublon with ui-ux-pro-max)
- python3 -c moved from deny to ask (unblocks graphify)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-12 14:56:22 +02:00

138 lines
5.6 KiB
Markdown

---
name: ship-feature
description: Ship feature end-to-end: design → plan → implement (TDD) → review → finish
argument-hint: <feature description>
disable-model-invocation: true
allowed-tools: Read, Write, Edit, Bash, Grep, Glob
---
# ORCHESTRATOR: SHIP FEATURE
## REQUEST
$ARGUMENTS
---
## STEP 0 — PLUGIN CHECK + AUTO-ACTIVATE
Load `$HOME/.claude/agents/plugin-advisor.md`. Feed request.
- ACTION REQUIRED → show RECOMMENDATIONS block, offer: A) fix plugins B) type "force". STOP.
- PROPOSED CHANGES exist → show list, ask "Apply? (yes / no / customize)". Apply on confirm.
- OK → `✅ Plugin check passed — [active plugins] — complexity: <score>%`, continue.
## STEP 0b — PROJECT CONTEXT CHECK
Verify the project has a `CLAUDE.md` and print a brief orientation summary:
```bash
ls CLAUDE.md .claude/CLAUDE.md 2>/dev/null | head -1
git branch --show-current 2>/dev/null || echo "not a git repo"
git log --oneline -3 --format="%h %<(50,trunc)%s" 2>/dev/null || true
ls .gsd/ROADMAP.md 2>/dev/null | head -1
```
- **CLAUDE.md found** → read it silently, then print orientation header (informational, not a gate):
```
📋 PROJECT CONTEXT
Project : <name from CLAUDE.md>
Stack : <stack from CLAUDE.md>
Branch : <current git branch>
Recent : <last 3 commit messages>
GSD : <current milestone if .gsd/ROADMAP.md exists, else "not initialized">
```
Continue to STEP 1.
- **Not found** →
Print: "⚠️ No CLAUDE.md found in this directory.
This project has not been onboarded into claude-config.
Run `/onboard` first to generate CLAUDE.md and project settings,
then re-run `/ship-feature`."
STOP.
## STEP 0c — CTX7 CACHE CHECK (if fast-libs in project)
Check if the project uses fast-evolving libs (scan `package.json` for next, react, prisma, supabase, drizzle, expo):
1. If `.ctx7-cache/` exists with recent files (<7 days old) print `📚 ctx7 cache found: <libs>` and continue.
2. If `.ctx7-cache/` missing or stale AND `ctx7` is installed AND fast-libs detected:
```bash
mkdir -p .ctx7-cache
# Fetch docs for each detected fast-lib (adapt to actual deps):
ctx7 docs /vercel/next.js "app router middleware routing" > .ctx7-cache/nextjs-core.md 2>/dev/null || true
ctx7 docs /prisma/prisma "schema client queries" > .ctx7-cache/prisma-core.md 2>/dev/null || true
```
Print: `📚 ctx7 docs pre-fetched for: <libs>`
3. If no fast-libs or `ctx7` not installed → skip silently.
During implementation (STEP 4), when making decisions about fast-lib APIs:
- Read the relevant `.ctx7-cache/<lib>.md` file before writing code.
- This avoids repeated ctx7 calls and keeps docs available without context cost.
## STEP 1 — BRAINSTORM
Invoke `superpowers:brainstorming`. Refine request into validated design via Socratic questioning. Don't proceed until design approved.
## STEP 2 — PLAN
Invoke `superpowers:writing-plans`. Break design into tasks (2-5 min each). Each task: exact file paths, full code, verification steps.
## STEP 3 — VALIDATION GATE ★ MANDATORY STOP
```
SHIP FEATURE — VALIDATION GATE
FEATURE: <n> | TASKS: <count>
<numbered task list>
Approve and execute? (yes / request changes)
```
Changes → back to STEP 2. Approved → continue.
## STEP 4 — IMPLEMENT
Invoke `superpowers:subagent-driven-development`. Isolated subagents. 2-stage review per task: spec compliance → code quality.
## STEP 4b — ERROR RECOVERY (if STEP 4 fails)
If a subagent returns a build error, failing test, or type error:
1. Load `$HOME/.claude/agents/analyzer.md` in DEBUG MODE on the exact error output.
Produce: root cause hypotheses (ordered), affected files, what NOT to touch.
2. Present gate:
```
SHIP FEATURE — ERROR IN STEP 4
TASK : <task name that failed>
ERROR : <one-line summary>
HYPOTHESES:
1. [HIGH] <cause> — evidence: <…>
2. [MED] <cause> — evidence: <…>
OPTIONS :
A) Apply fix for hypothesis 1 and re-run this task
B) Skip this task and continue with remaining tasks
C) Abort feature — preserve work done so far
```
3. Wait for user choice. Do NOT auto-fix. Do NOT proceed without explicit approval.
4. If A → apply minimal fix, re-run STEP 4 for the failed task only. Max 2 retry attempts.
If still failing after 2 → fall back to options B or C.
If B → before skipping: scan remaining task list for tasks that depend on the failed task
(look for references to the same file or function in subsequent tasks).
If dependents found → present: "Tasks [N, M] depend on the skipped task.
Skip them too? (yes / keep and accept partial implementation)"
If no dependents → skip cleanly and continue.
## STEP 5 — ANALYZE
Load `$HOME/.claude/agents/analyzer.md`. Check: no regressions, no stale code, no plan deviations.
## STEP 6 — CODE REVIEW
Invoke `superpowers:requesting-code-review`. Fix all CRITICAL before proceeding.
## STEP 7 — FINISH
Invoke `superpowers:finishing-a-development-branch`. Tests pass, build clean, ready to merge.
## STEP 8 — SYNC README
Load `$HOME/.claude/agents/readme-updater.md` with arg `sync`. Update cmds/vars/structure, add recent changes entry.
---
## RULES
- No skipping steps. No merged agent responsibilities.
- No implement without user approval at STEP 3.
- Subagents isolated — no shared context between tasks.
- Fix all CRITICAL review issues before proceeding.
- Stop if requirements unclear at any step.
- STEP 4 errors → STEP 4b gate required before any fix. Never auto-patch a failing subagent.
---
## FINAL OUTPUT
```
FEATURE SHIPPED: <n>
TASKS: <N>/<N> | TESTS: ✅/❌ | REVIEW: APPROVED/CHANGES REQUIRED
REMAINING ISSUES: <list or none>
```