commit
Skill75 starsChanged 2 months agoMITView on GitHub
Stages, commits, and optionally pushes local changes with a Conventional Commits message — always previews and confirms first, and offers a feature branch on protected branches. Modes: "suggest" proposes a message without committing (read-only); "branch [description]" creates and switches to a conventionally named branch, never committing or altering local changes. Multi-repo aware.
- Reads credentials
- Commits and pushes
188 other projects use a version of this.
Adopted by
Most recently taken up by
- daledesilva/obsidian_inktheir own version, first seen Nov 9, 2026
- thoughtbot/rails-consultanttheir own version, first seen Nov 8, 2026
- pedrohcgs/claude-code-my-workflowtheir own version, first seen Nov 8, 2026
- jisuzzzz/haechitheir own version, first seen Nov 8, 2026
- dirien/yet-another-agent-harnesstheir own version, first seen Nov 8, 2026
- denysdovhan/skillstheir own version, first seen Nov 8, 2026
Used with
Projects with it also use validate-plan, act-on-feedback and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
shawn-dumasSkillvalidate-plan
Validate an orchestration plan after pre-flight certification. Runs adversarial review, deep review (verify prompt data against codebase), and PoC gate for risky approaches. Auto-invoked by orchestrate-* skills at Step 8.
Also in 5 projects that use this
MicrosoftSkillact-on-feedback
Act on user feedback attached to the current session. Use when the user submits feedback on the session's changes via the Submit Feedback button.
Also in 3 projects that use this
decocmsSkilladd-mcp-tools
Guide for adding new MCP tools with consistent patterns for schemas, tool definitions, registry updates, and Better Auth integration
Also in 3 projects that use this
MicrosoftSkillgenerate-run-commands
Generate or modify run commands for the current session. Use when the user wants to set up or update run commands that appear in the session's Run button.
Also in 3 projects that use this
mitsuhikoSkilloebb-scotty
Austrian rail travel planner (ÖBB Scotty). Use when planning train journeys in Austria, checking departures/arrivals at stations, or looking for service disruptions. Covers ÖBB trains, S-Bahn, regional trains, and connections to neighboring countries.
Also in 3 projects that use this
mitsuhikoSkillpi-share
Load and parse session transcripts from shittycodingagent.ai/buildwithpi.ai/buildwithpi.com/pi.dev (pi-share) URLs. Fetches gists, decodes embedded session data, and extracts conversation history.
Also in 3 projects that use this
Other versions188 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 188
- microsoft/vscode★ 195ksame name
- TryGhost/Ghost★ 56ksame name
- ToolJet/ToolJet★ 41ksame name
- ShareX/ShareX★ 41ksame name
- openai/symphony★ 27ksame name
- ccusage/ccusage★ 19ksame name
- ryoppippi/ccusage★ 19ksame name
- windmill-labs/windmill★ 18ksame name
- EverMind-AI/EverMemOS★ 14ksame name
- EverMind-AI/EverOS★ 14ksame name
178 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository20 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .claude/CLAUDE.mdCLAUDE.md
- brainstormSkill
- code-reviewSkill
- deepSkill
- dreamSkill
- gauntletSkill
- handoffSkill
- how-to-runSkill
- initSkill
- jiraSkill
- paper-initSkill
- permissionsSkill
- promptSkill
- prSkill
- refactorSkill
- resetSkill
- tddSkill
- unit-testSkill
- worktreeSkill
The file
We split this skill into the rules it gives an agent
11 rules. 0 are common, 5 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 168 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 17
Untracked-file gate: if git status --short shows untracked files (??):
- Distinctive: line 20
Read the contents of the included untracked files — step 1 gathered only their names.
- Very specific: line 22
Generate the message: read $CLAUDE_PLUGIN_ROOT/skills/commit/references/conventional-commit-format.md and apply it to everything gathered.
- Distinctive: line 27
Not protected: "Commit and push" / "Commit only" / "Edit message" / "Cancel"
- Very specific: line 32
If a "Create branch" option was chosen: git checkout -b <name>.
- Distinctive: line 34
Commit with a heredoc to preserve the multi-line message:
- Distinctive: line 43
If the commit fails, report the error and stop — never proceed to push.
- Distinctive: line 52
Never stages, commits, or modifies anything.
- Very specific: line 55
Read $CLAUDE_PLUGIN_ROOT/skills/commit/references/conventional-commit-format.md and generate the message(s).
- Very specific: line 75
Name and create: read $CLAUDE_PLUGIN_ROOT/skills/commit/references/branch-naming.md.
- Very specific: line 86
When the work is ready to commit, run /optimus:commit in this conversation so the implementation context is captured.
“Common” is measured: kept by at least 6 in 10 of the 168 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Commit
Pick the mode from the arguments: suggest → Suggest mode; branch (optionally followed by a description) → Branch mode; anything else → Default mode.
Multi-repo: if git rev-parse --is-inside-work-tree does not return true, read $CLAUDE_PLUGIN_ROOT/skills/init/references/multi-repo-detection.md and apply it. When it returns true, resolve the repository root with git rev-parse --show-toplevel, including in a linked worktree or subdirectory. In a workspace, git commands run inside each child repo; Default and Suggest modes process each repo with changes independently (per-repo steps, per-repo preview under a ## <repo-name> heading, one combined final summary); Branch mode targets a single repo (its step 1).
Default mode — stage, commit, optionally push
- Gather changes:
git diff --cached,git diff,git status --short(untracked = names only), leading with--statwhen the diffs are large. In a workspace run these inside each child repo and skip the clean ones. No local changes anywhere → say so and stop. - Untracked-file gate: if
git status --shortshows untracked files (??):- List them and warn about any that look like secrets (
.env,*.key,*.pem,*.pfx,credentials.*,secrets.*,*.sqlite,*.db). AskUserQuestion— header "Untracked files": "Include all" / "Exclude all" / "Let me choose". "Include all" never covers secret-looking files — each one requires individual confirmation.- Read the contents of the included untracked files — step 1 gathered only their names.
- List them and warn about any that look like secrets (
- Generate the message: read
$CLAUDE_PLUGIN_ROOT/skills/commit/references/conventional-commit-format.mdand apply it to everything gathered. If changes span multiple concerns,AskUserQuestionwhether to commit together or split; when splitting, run steps 4–6 once per commit with its pre-assigned files. - Protected-branch check: look for
.claude/hooks/restrict-paths.sh— in a workspace check the child repo first, then the workspace root (first found wins). Parse itsPROTECTED_BRANCHESarray to decide whether the current branch is protected; no hook = unprotected; hook present but array unparseable = treat as unprotected and note it in the preview. If protected, generate a feature branch name per$CLAUDE_PLUGIN_ROOT/skills/commit/references/branch-naming.md—<type>from the commit message, slug from its subject line. - Preview and confirm: show the branch, the full commit message (subject and body — never truncate to subject-only), and the files to stage.
AskUserQuestion— "How would you like to proceed?":- Not protected: "Commit and push" / "Commit only" / "Edit message" / "Cancel"
- Protected: "Create branch
<name>, commit, and push" / "Create branch<name>and commit only" / "Edit message" / "Cancel" - "Edit message" takes the user's adjusted message and re-presents this step (on a protected branch, regenerate the feature branch name from it). "Cancel" aborts with zero repo changes.
- Execute:
- If a "Create branch" option was chosen:
git checkout -b <name>. If creation fails (name taken), report it and let the user pick a different name or cancel — never silently auto-suffix a name the user approved (deliberate divergence from branch-naming.md's Collision Handling). - Stage the specific files from steps 1–2 — never
git add -A, and never a secret-looking file unless individually confirmed in step 2. Commit with a heredoc to preserve the multi-line message:
git commit -m "$(cat <<'EOF' <message> EOF )"- If the commit fails, report the error and stop — never proceed to push.
- Push only if the user chose a push option; use
git push -u origin <branch>when no upstream exists.
- If a "Create branch" option was chosen:
- Report: created branch (if any),
Committed: <short-hash> <subject>per repo,Pushed to: origin/<branch>if pushed. If a feature branch was created, tell the user they are now on it.
Recommend /optimus:pr when a pull request is next — stay in this conversation so the implementation context is captured.
Suggest mode — message only, read-only
Never stages, commits, or modifies anything.
- Gather changes as in Default mode step 1.
- Read
$CLAUDE_PLUGIN_ROOT/skills/commit/references/conventional-commit-format.mdand generate the message(s). - Present each message in a copyable code block. If changes span multiple concerns, propose separate commits, each with its message and the exact files to stage. In a workspace, put each repo's suggestion under a
## <repo-name>heading (label even a single repo).
Recommend /optimus:commit to actually commit — stay in this conversation so the implementation context is captured.
Branch mode — create a named branch, nothing else
A purely local move: git checkout -b only. Never commit, push, stage, stash, reset, or modify any file or the index — staged, unstaged, and untracked changes carry to the new branch untouched. Ask questions only for multi-repo ambiguity.
- Target repo (workspace only): one repo with local changes → target it silently, unless the description or conversation names a different repo (that repo's clean tree is the starting-fresh case; leave the dirty repo untouched). Multiple dirty repos →
AskUserQuestion— header "Target repo", one option per repo. No dirty repos → pick the repo the description or conversation points to; ask if ambiguous; with no context at all, inform the user and stop. Derive the description: record the current branch for the report. Use the first source with enough signal: (1) inline description from the arguments, (2) conversation context, (3) analysis of local changes, including untracked files. If none provides enough signal, print exactly:
Could not determine a meaningful branch name from the conversation or local changes. Provide a description, e.g., `/optimus:commit branch "add user authentication"`and stop — never create a branch with a generic or meaningless name.
Name and create: read
$CLAUDE_PLUGIN_ROOT/skills/commit/references/branch-naming.md; apply its Type Detection Keywords, Slug Rules, and Collision Handling. Thengit checkout -b <branch-name>and report:## Branch Created `<branch-name>` from `<original-branch>`. Local changes preserved (nothing committed or pushed).Omit the last line when the tree was clean; in a workspace, name the target repo.
When the work is ready to commit, run /optimus:commit in this conversation so the implementation context is captured.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

