skills/describe_pr/SKILL.md
Generate comprehensive PR descriptions following repository templates
npx skillsauth add rubicanjr/FinCognis skills/describe_prInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
You are tasked with generating a comprehensive pull request description following the repository's standard template.
Read the PR description template:
thoughts/shared/pr_description.md existsthoughts/shared/pr_description.mdIdentify the PR to describe:
gh pr view --json url,number,title,state 2>/dev/nullgh pr list --limit 10 --json number,title,headRefName,authorCheck for existing description:
thoughts/shared/prs/{number}_description.md already existsGather comprehensive PR information:
gh pr diff {number}gh repo set-default and select the appropriate repositorygh pr view {number} --json commitsgh pr view {number} --json baseRefNamegh pr view {number} --json url,title,number,state4b. Gather reasoning history (if available):
ls .git/claude/commits/*/reasoning.md 2>/dev/nullbash "$CLAUDE_PROJECT_DIR/.claude/scripts/aggregate-reasoning.sh" mainAnalyze the changes thoroughly: (ultrathink about the code changes, their architectural implications, and potential impacts)
Handle verification requirements:
make check test, npm test, etc.), run it- [x]- [ ] with explanationGenerate the description:
Save the description:
thoughts/shared/prs/{number}_description.mdUpdate the PR:
gh pr edit {number} --body-file thoughts/shared/prs/{number}_description.mddevelopment
Goal-based workflow orchestration - routes tasks to specialist agents based on user goals
tools
Wiring Verification
development
Connection management, room patterns, reconnection strategies, message buffering, and binary protocol design.
development
Screenshot comparison QA for frontend development. Takes a screenshot of the current implementation, scores it across multiple visual dimensions, and returns a structured PASS/REVISE/FAIL verdict with concrete fixes. Use when implementing UI from a design reference or verifying visual correctness.