Use when /proctor needs to convert a GitHub PR (metadata + diff) into a structured ChangeMap categorizing every changed hunk. First stage of the PRoctor pipeline. Output is a single JSON object — no prose. Use when you see "apply skill analyzing-pr-changes" or when the orchestrator hands you pr.json + diff.patch.
EnglishUse after the approval gate to dispatch each item in an ApprovedPlan to the pr-test-executor subagent and assemble TestResults. Third stage of the PRoctor pipeline. Output is a single TestResults JSON object — no prose.
EnglishUse after executing-pr-tests when there are failures and autofix: true. Dispatches the pr-test-fixer subagent for each failure. In CI mode (PROCTORPUSHFIX=1) applies patches on a fix branch and opens a fix PR; in local mode writes patches to disk for the developer to review and apply manually. Returns a FixPRRef. Fourth stage of the PRoctor pipeline. Output is a single FixPRRef JSON object (or null if nothing to fix).
EnglishUse when /proctor has a ChangeMap and needs to produce a concrete TestPlan — one test item per behavior worth verifying. Second stage of the PRoctor pipeline. Output is a single JSON object — no prose. Use when handed change-map.json and optionally .pr-test.yml.
EnglishFinal stage of /proctor. Renders TestResults + FixPRRef into a markdown report. In CI mode (PROCTORPOSTCOMMENT=1) posts the comment to the PR; in local mode saves the report to disk and prints the path so the developer can read it locally. Each item gets its own section with evidence, command, output excerpt, logs, and screenshot refs (when present).
English
AGENT SKILLS ATLASIncludes Analyzing PR Changes, Executing PR Tests, Fixing Test Failures, and other workflows.
Claude Code
Codex
Cursor
GitHub Copilot
Gemini CLISupported by the Atlas install prompt. Skill-specific requirements vary.
Pack stats
What’s included
Analyzing PR Changes
Use when /proctor needs to convert a GitHub PR (metadata + diff) into a structured ChangeMap categorizing every changed hunk. First stage of the PRoctor pipeline. Output is a single JSON object — no prose. Use when you see "apply skill analyzing-pr-changes" or when the orchestrator hands you pr.json + diff.patch.
Executing PR Tests
Use after the approval gate to dispatch each item in an ApprovedPlan to the pr-test-executor subagent and assemble TestResults. Third stage of the PRoctor pipeline. Output is a single TestResults JSON object — no prose.
Fixing Test Failures
Use after executing-pr-tests when there are failures and autofix: true. Dispatches the pr-test-fixer subagent for each failure. In CI mode (PROCTORPUSHFIX=1) applies patches on a fix branch and opens a fix PR; in local mode writes patches to disk for the developer to review and apply manually. Returns a FixPRRef. Fourth stage of the PRoctor pipeline. Output is a single FixPRRef JSON object (or null if nothing to fix).
Planning PR Tests
Use when /proctor has a ChangeMap and needs to produce a concrete TestPlan — one test item per behavior worth verifying. Second stage of the PRoctor pipeline. Output is a single JSON object — no prose. Use when handed change-map.json and optionally .pr-test.yml.
Reporting PR Test Results
Final stage of /proctor. Renders TestResults + FixPRRef into a markdown report. In CI mode (PROCTORPOSTCOMMENT=1) posts the comment to the PR; in local mode saves the report to disk and prints the path so the developer can read it locally. Each item gets its own section with evidence, command, output excerpt, logs, and screenshot refs (when present).
Quick start
- Choose your skills. Select individual skills above, or choose the entire pack.
- Copy the install prompt. Use Copy install prompt above the list.
- Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.
Try it with your agent
After installation, start with a request like:
Use the Analyzing PR Changes skill to help with my task.
Ask me for the context you need before getting started.Requirements & license
- A coding agent that can access your project files and install skills.
- Your agent checks each skill’s instructions, dependencies, and permissions during setup.
- Reported license for the skills shown: MIT. The author’s terms apply.