humansys/raise/rai-bugfix-fix
Claude Code and agent skill by humansys/raise
- community
Checking that the SKILL.md file is still in the repository.
Default variant: TDD + gates + quality-review. 11 steps vs 19 enterprise, no HITL per-task.
Install
npx vskill@latest install humansys/raise/rai-bugfix-fixSKILL.md
Bugfix Fix
ℹ DEFAULT VARIANT (ADR-134 v2) — si cambias lógica [R]/[G] en rai-bugfix-fix, propaga aquí.Fuente: work/epics/e14770-lean-methodology/skill-audit.md §rai-bugfix-fixLean residue: 11 pasos [R9, G1, I1] de 19 en enterprise (42% ceremonia eliminada).Eliminado: HITL per-task pause (paso 9 del enterprise), bloques verification/if-blocked, checklist.
Objetivo
Ejecutar el fix con TDD estricto. Sin esperar acknowledgment per-task — ejecución autónoma (Ri mode). La calidad la garantizan los gates y el quality-review al final.
Inputs: Bug ID, scope.md (producido por diagnose). Expected state: En bug branch, árbol limpio.
Excepción — Path A (bugfix epic-scoped, RAISE-15901): si el commit inicial de scope.md incluye trailer Epic:, rai-bugfix-start corrió Path A y este fix vive sobre epic/*, no sobre bug/* — es el estado esperado para ese caso, no un working tree inesperado.
Por cada task del scope
0. Emitir señal de inicio
1. RED [R]
Escribir test fallido que define el comportamiento esperado.
2. GREEN [R]
Escribir código mínimo que hace pasar el test.
3. REFACTOR [R]
Limpiar mientras los tests siguen verdes.
4. Ejecutar 4 gates [R]
Environment setup (mandatory — eliminates RC2 path errors, RAISE-15430 AC2): Use absolute paths for all file operations: $WORKTREE_ROOT/<relative-path>.
Interface lookup (RC1, RAISE-15431) — query before you guess. Before invoking any rai <cmd> subcommand or raise_cli import you have NOT verified this session:
- Query the graph:
rai graph query "<cmd> <subcommand>" --types symbol --limit 3
- For Jira transitions/priorities:
rai graph query "jira transition workflow" --types pattern(PAT-F-1826 has the RAISE workflow; PAT-E-9996..9998 have priority/issue-type/transition values) - For SCM tools (glab, gh):
rai graph query "glab/gh <cmd> <subcommand>" --types pattern --limit 3(glab/gh have no Python symbols — PAT-E-10000..10009 cover flags, 409/404 errors, glab vs gh distinctions) - For SQLite/raise.db schema:
rai graph query "raise.db <table> schema" --types pattern --limit 3(schema is not in code symbols — PAT-E-9999..10007 have column lists)
- Fallback if 0 results:
rai <cmd> --help| for SCM tools:glab/gh <cmd> --help| for DB:sqlite3 db ".schema"
Skip when the interface was already verified this session. One query beats a spiral of guessed-flag retries (70% of RC1 errors spiral).
Si un gate falla → fijar y re-verificar (máx 3 intentos). Si persiste → STOP: documentar blocker, escalar.
5. Commit por task [G]
6. Si hay más tasks → volver al paso 1 [R]
7. Quality review de la implementación [R]
Invocar /rai-quality-review sobre los archivos cambiados.
| Veredicto | Acción | |
|---|---|---|
| PASS / PASS WITH RECOMMENDATIONS | Continuar | |
| FAIL | Atender findings críticos, re-run gates, continuar |
8. Emitir outcome + señal de cierre
Emitir bloque outcome antes de devolver control:
STOP. Devolver: commits realizados, veredicto QR, gate status.
Backlog tracker
Engine behavior (RAISE-16987): The engine checks ExecutionContext to determine mode:
mode: pipeline— engine active;apply_phase_transitionmoves the tracker. No skill call required.mode: standalone— no active run; tracker is not moved automatically.
If standalone (mode == standalone via ExecutionContext):
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingpackages/raise-cli/src/raise_cli/skills_base/rai-bugfix-fix/SKILL.md
Popularity
- Repository stars
- 71
- Repository forks
- 8
- 7-day trend score
- 21
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/humansys/raise/rai-bugfix-fix)