Run QA checks and produce the QA report.
Scanned 2/12/2026
Install via CLI
openskills install GrinRus/ai_driven_dev---
name: qa
description: Run QA checks and produce the QA report.
argument-hint: $1 [note...]
lang: ru
prompt_version: 1.0.31
source_version: 1.0.31
allowed-tools:
- Read
- Edit
- Glob
- "Bash(${CLAUDE_PLUGIN_ROOT}/skills/qa/scripts/preflight.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/skills/qa/scripts/run.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/skills/qa/scripts/postflight.sh:*)"
- "Bash(rg:*)"
- "Bash(npm:*)"
- "Bash(pnpm:*)"
- "Bash(yarn:*)"
- "Bash(pytest:*)"
- "Bash(python:*)"
- "Bash(go:*)"
- "Bash(mvn:*)"
- "Bash(make:*)"
- "Bash(./gradlew:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/rlm-slice.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/set-active-stage.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/qa.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/tasks-derive.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/progress.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/stage-result.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/status-summary.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/tasklist-check.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/tasklist-normalize.sh:*)"
- "Bash(${CLAUDE_PLUGIN_ROOT}/tools/set-active-feature.sh:*)"
model: inherit
disable-model-invocation: true
user-invocable: true
context: fork
agent: qa
---
Follow `feature-dev-aidd:aidd-core` and `feature-dev-aidd:aidd-loop`.
## Steps
1. Preflight reference: `skills/qa/scripts/preflight.sh`. This step is mandatory and must produce `readmap/writemap`, actions template, and `stage.preflight.result.json`.
2. Read order after preflight: `readmap.md` -> loop pack -> review pack (if exists) -> rolling context pack; do not perform broad repo scan before these artifacts.
3. Run QA via `${CLAUDE_PLUGIN_ROOT}/tools/qa.sh` and derive tasks if needed.
4. Fill actions.json (v1): create `aidd/reports/actions/<ticket>/<scope_key>/qa.actions.json` from template and validate schema via `skills/qa/scripts/run.sh` before postflight.
5. Postflight reference: `skills/qa/scripts/postflight.sh`. Apply actions via DocOps, then run progress check, stage-result, status-summary.
## Notes
- QA stage runs full tests per policy.
No comments yet. Be the first to comment!
Practical guide to testing web applications with screen readers for comprehensive accessibility validation.
克劳德代码会话的正式评估框架,实施评估驱动开发(EDD)原则
Go测试模式包括表格驱动测试、子测试、基准测试、模糊测试和测试覆盖率。遵循TDD方法论,采用地道的Go实践。
使用pytest、TDD方法、夹具、模拟、参数化和覆盖率要求的Python测试策略。
在编写新功能、修复错误或重构代码时使用此技能。强制执行测试驱动开发,包含单元测试、集成测试和端到端测试,覆盖率超过80%。