From claude-skills
Finalize a feature — run tests, build, and present options for committing, creating a PR, or discarding. Use when user says "finish", "wrap up", "done", "create PR", "merge", or wants to finalize their work on a feature.
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-skills:finish-featureThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Verify the feature is ready, then help the user decide what to do with it.
Verify the feature is ready, then help the user decide what to do with it.
Run these checks in order:
Feature tests:
pnpm vitest run src/features/{feature}/
Report: X tests passed, Y failed
Build:
pnpm build
Report: success or failure with error summary
Uncommitted changes:
git status
Report: X files modified, Y untracked
If tests or build fail, report the failures and ask the user:
Look for PROGRESS.md files to determine the base branch:
PROGRESS.md in the feature directory (src/features/{feature}/)**Base Branch** field if it existsAlso check .claude/pipeline/{feature}/ for any pipeline artifacts.
Use AskUserQuestion to present:
git diff --stat and a summary of what was builtgh pr creategit status and git diff --stat to see what's changedpnpm build before each commit to verifyCo-Authored-By linesgit push -u origin {branch-name}
gh pr create --base {base-branch} --title "{title}" --body "{body}"
Base Branch field, or ask user if not foundThis is destructive. Always confirm:
"This will discard ALL uncommitted changes on this branch. Are you sure? This cannot be undone."
Only proceed if user explicitly confirms.
After committing or creating a PR:
.claude/pipeline/{feature}/DESIGN.md (brainstorm output)*-implementation-plan.md execution PROGRESS.mdpnpm build before committing — catch type errors earlynpx claudepluginhub longjohnsilver1504/claude-skillsProvides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.