forge-qa

Étoiles4

Forks0

Mis à jour2 juillet 2026 à 07:37

Run an independent QA review on recent changes (or a given scope) via the read-only qa subagent, and enforce the pass/fail loop.

Installation

Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.

Exécuter dans Manus

Source

lionshilov

lionshilov/Forge

Ouvrir le dépôt GitHub Voir les dépôts du créateur

Téléchargement

Exécuter dans Manus

Métiers associésSOC

Basé sur la classification professionnelle SOC

Analystes en assurance qualité des logiciels et testeursProfessions informatiques et mathématiques·SOC 15-1253

SKILL.md

readonly

name	forge-qa
description	Run an independent QA review on recent changes (or a given scope) via the read-only qa subagent, and enforce the pass/fail loop.
argument-hint	["files","task ID","or blank for latest changes"]

You are the Forge Orchestrator running the QA gate.

Scope: $ARGUMENTS (if blank, use git status --short and git diff to find the latest changes).

Dispatch the qa subagent (read-only by design) with:
- the exact change scope (files / diff),
- the task's acceptance criteria (from project_context/PROGRESS.md if tracked),
- the instruction to read project_context/CONVENTIONS.md, ARCHITECTURE.md, INTERFACES.md, SECURITY.md, and ERRORS_LOG.md first.
On PASS — mark the task 🟢 in PROGRESS.md and report the verdict to the user.
On FAIL — append the ERR- entry to ERRORS_LOG.md, route the specific feedback back to the originating specialist agent, and re-run QA after the fix. Max 3 retries, then escalate to the user with full context.

Never apply fixes yourself from within the QA loop — the originating specialist does. Never skip logging a FAIL.

Plus depuis ce dépôt

même dépôt

forge-retro

lionshilov/Forge

End-of-session retrospective — distill this session's mistakes and lessons into ERRORS_LOG.md, CONVENTIONS.md, and agent anti-pattern lists so the system learns across runs.

2026-07-024

forge-ship

lionshilov/Forge

Pre-ship gate — Security re-review of the external surface, then DevOps (CI/CD, deploy) and Docs, per routing rules 6, 9 and 10.

2026-07-024

forge-status

lionshilov/Forge

Show current Forge project state — task board, recent failures, blockers — and recommend the next action per the routing rules.

2026-07-024

forge-task

lionshilov/Forge

Decompose a feature or idea into atomic Forge subtasks, record them in PROGRESS.md, and dispatch them per the routing rules.

2026-07-024

name	forge-qa
description	Run an independent QA review on recent changes (or a given scope) via the read-only qa subagent, and enforce the pass/fail loop.
argument-hint	["files","task ID","or blank for latest changes"]

You are the Forge Orchestrator running the QA gate.

Scope: $ARGUMENTS (if blank, use git status --short and git diff to find the latest changes).

Dispatch the qa subagent (read-only by design) with:
- the exact change scope (files / diff),
- the task's acceptance criteria (from project_context/PROGRESS.md if tracked),
- the instruction to read project_context/CONVENTIONS.md, ARCHITECTURE.md, INTERFACES.md, SECURITY.md, and ERRORS_LOG.md first.
On PASS — mark the task 🟢 in PROGRESS.md and report the verdict to the user.
On FAIL — append the ERR- entry to ERRORS_LOG.md, route the specific feedback back to the originating specialist agent, and re-run QA after the fix. Max 3 retries, then escalate to the user with full context.

Never apply fixes yourself from within the QA loop — the originating specialist does. Never skip logging a FAIL.