From devkit
Stage and create a git commit with a consistent message format. Use when the user says "commit this", "make a commit", or "write a commit message". Mirrors the title convention used by the `write-pr` skill so commits and PRs read consistently.
How this skill is triggered — by the user, by Claude, or both
Slash command
/devkit:commitThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Create a git commit with a clear, specific message grounded in the actual diff. Never commit without the user's explicit go-ahead in this session.
Create a git commit with a clear, specific message grounded in the actual diff. Never commit without the user's explicit go-ahead in this session.
Run these in parallel before drafting a message:
git status (no -uall) — see what's staged, unstaged, untracked.git diff --staged and git diff — read both. Draft messages from the diff, not from memory.git log -n 5 --oneline — match the repo's existing tone if it diverges from this convention.If nothing is staged, decide with the user whether to stage everything, stage specific files, or abort. Never silently git add -A — it can sweep in .env, large binaries, or unrelated work.
Subject line: [type][Domain] Summary
feat, fix, refactor, perf, test, chore, docsAuth, API, CI)Body (when the change isn't self-evident from the subject):
Skip the body for trivial changes (typo fixes, dependency bumps, single-line tweaks).
--amend a previously-pushed commit unless the user explicitly asks.--no-verify. If a hook fails, fix the underlying issue and create a new commit — do not modify the failed one..env, credentials.*, key files). Warn loudly if the user insists.Always pass the message via heredoc to preserve formatting:
git commit -m "$(cat <<'EOF'
[type][Domain] Summary
Optional body explaining why.
EOF
)"
After committing, run git status to confirm a clean tree.
Provides 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.
npx claudepluginhub markusjohansen/mj-plugins --plugin devkit