菜单
Scans GitHub for open issues labeled 'approved' but not 'in progress', then processes each sequentially via fix-issue. Runs hourly.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Audits each directory in ./src for bugs, security vulnerabilities, and performance issues sequentially. Generates one consolidated issue per directory via /create-issue.
Audits Agent Skills SKILL.md files against the protocol specification and best practices. Identifies issues across 10 categories: schema violations, naming, description quality, structure, progressive disclosure, file references, best practices, calibration, content quality, and optional directories. Produces a structured report with severity levels and actionable fixes.
Expert System Prompt Evaluator Agent. Analyzes, rates, and provides actionable feedback for system prompts against a 7-criteria framework. Returns structured machine-readable JSON output with weighted scores, evidence-based reasoning, and harness-aware recommendations.
Automates the complete git workflow: scans project rules for conventions, stages all changes, commits, pushes to the remote, and opens a Pull Request. Ensures strict compliance with documentation rules, including rule 5.4.
Orchestrates a full feature lifecycle: receive goals, synthesize detailed specs, propose via OpenSpec, commit & push, apply tasks, audit results, update PR, and post audit results as a comment.
Receives a user description, synthesizes it into a title and description, categorizes as 'fix' or 'feat', creates a GitHub issue, audits the codebase for actionable details, and updates the issue with findings.
| name | scan-issues |
| description | Scans GitHub for open issues labeled 'approved' but not 'in progress', then processes each sequentially via fix-issue. Runs hourly. |
| license | BSD 3-Clause |
| compatibility | Requires gh CLI authenticated with the avoidwork/madz repo. Must be run from the project root. |
| metadata | {"agent":"coding"} |
An autonomous issue scanner and fixer. Finds approved, unassigned issues and processes them one at a time through the fix-issue pipeline.
gh issue list --state open --label approved --json number,title,url,labels --repo avoidwork/madz
This fetches all open issues that have the approved label. Note: The in progress label check is handled by fix-issue in its Step 3, so we don't filter here — let each issue's own validation decide.
If no issues match, report No approved, unassigned issues found. and exit cleanly.
Extract each issue's number, title, url, and labels from the JSON array. Sort the remaining issues by number ascending (oldest first) to ensure sequential processing.
Process one issue at a time. Do NOT spawn subagents or parallelize — each issue must be processed sequentially to avoid label conflicts and race conditions:
For each issue in the sorted list:
fix-issue <ISSUE_NUMBER>
Rate limiting: If GitHub API returns 403 rate limit, wait 60 seconds and retry the current issue once.
Timeout handling: Each fix-issue run can take 30-60 minutes (via create-feature). If a run exceeds 60 minutes, report a timeout warning but do not abort — the issue may still be processing.
After all issues are processed, produce a structured summary:
Scan complete.
✅ Fixed: #42 — "Crash on empty input"
✅ Fixed: #57 — "Missing error boundary"
❌ Skipped: #89 — "Auth timeout race" (fix-issue failed)
Total scanned: 3
Success: 2
Failed: 1
gh auth: Suggest gh auth login.This skill is designed to run hourly. When triggered via cron, execute the full workflow above without requiring user input.