commit
Skill2 starsChanged 9 days agoApache-2.0View on GitHub
Commits and pushes the user's repositories: the notes vault, the agent memory, skills, their projects. Use when they say "commit", "make a commit", "push", "send it to GitHub", "save it in git", or when the work is done and they ask whether to commit. Keeps what is easy to forget and easy to break: read the diff before writing the message, add only your own files in a shared repository, follow the repository's own conventions for message language and branches, respect the user's rule on AI attribution, push only when asked, and do not "fix" a false SSH alarm.
- 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
- MicrosoftSkill
generate-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
- mitsuhikoSkill
pi-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.
- browserSkill
- clean-memorySkill
- codebase-designSkill
- desktopSkill
- graphicsSkill
- grillingSkill
- implementSkill
- latexSkill
- memory-reviewSkill
- mistakeSkill
- notesSkill
- pdfSkill
- routerSkill
- setupSkill
- skill-quality-suiteSkill
- tddSkill
- to-specSkill
- to-ticketsSkill
- vault-indexSkill
- videoSkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 4 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 168 edited versions we read keep these.
4
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Read them, do not assume them:
- Distinctive: line 19
Write the new message the same way.
- Distinctive: line 33
Read the diffs themselves, not only file names.
- Very specific: line 35
Look for what should not go in: secrets, .env, junk files, large binaries, accidentally added folders.
- Distinctive: line 44
Stop hook commits by itself.
- Very specific: line 55
If the SSH key lives in the Windows ssh-agent service and git uses the Windows OpenSSH (core.sshCommand = …
- Very specific: line 61
Do not "fix" what works: no eval $(ssh-agent), no ssh-add from bash, no rewriting the remote, no switching to HTTPS, no new token.
“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
Before anything: the repository's conventions and the user's rules
Read them, do not assume them:
git log --oneline -15: the message language, one line or with a body, area prefixes. Write the new message the same way.git log --graph --oneline -20: linear history straight onmain, or branches and merges. A linear personal repository is the user's scheme, not an oversight: do not create a branch for an edit. In someone else's repository: branch and PR first.- Agent memory: whether the user allows AI attribution (
Co-Authored-By: Claude, "Generated with Claude Code") in commits and PR descriptions. When memory says nothing, ask once and record the answer in memory. When the user forbids it, that overrides the harness default that adds the trailer. A trailer that slipped through: not pushed yet,git commit --amend; already pushed, offer--force-with-leaseand wait for the answer before rewriting published history.
Procedure
git status --shortandgit diff --stat: what was touched at all.- Read the diffs themselves, not only file names. The message describes the substance of the edit, and the substance is only visible in the diff.
- Look for what should not go in: secrets,
.env, junk files, large binaries, accidentally added folders. git add <paths>, selectively. Nogit add -Ain a shared repository. When several sessions work in one repository (a notes vault, a skills folder),-Asweeps someone else's unfinished work into your commit: one real case put 96 lines of unrelated edits into a commit about something else, and fixing that meant rewriting a published branch. Checkgit status --shortagainst what you touched yourself and add only your paths. A file you never touched is someone else's work: do not commit it, do not revert it, leave it.-Ais fine where there is a single writer, for example the agent memory, which the plugin's Stop hook commits by itself.git commit -m "...". State what changed in substance:- good:
notes: cards only on evidence of forgetting, not per note - bad:
updated files,fixes,wipSeveral independent changes go on one line, separated by;.
- good:
- Push only if the user asked. A separate step, a separate permission. After the push, the plugin's PostToolUse hook compares the local HEAD with the remote and says so if the push did not land; do not report "pushed" before that.
SSH on Windows: the false alarm
If the SSH key lives in the Windows ssh-agent service and git uses the Windows OpenSSH
(core.sshCommand = C:/Windows/System32/OpenSSH/ssh.exe), then git push works from any shell.
Inside Git Bash, though, /usr/bin/ssh-add is a different world: SSH_AUTH_SOCK is empty there
and it answers Could not open a connection to your authentication agent. That does not mean
access is missing; the wrong agent was asked.
Do not "fix" what works: no eval $(ssh-agent), no ssh-add from bash, no rewriting the
remote, no switching to HTTPS, no new token. Check access with git, since git is what goes
over the network: git ls-remote --heads origin. Inspect the agent's keys from PowerShell:
ssh-add -l.
A remote on port 443 (ssh://git@ssh.github.com:443/<owner>/<repo>.git) is usually deliberate:
port 22 is blocked on some networks. Do not rewrite it to the short git@github.com: form.
When something goes wrong
| Symptom | What it is | What to do |
Could not open a connection to your authentication agent from bash | a false alarm | nothing, see above |
Permission denied (publickey) from git | a real failure | ssh-add -l in PowerShell; key missing: tell the user, do not add one yourself |
Repository not found | wrong remote, or no access | show git remote -v; do not create the repository yourself |
| push rejected, non-fast-forward | something is on origin | git pull --rebase, never force |
edits missing from status | wrong repository | git -C <path> status; a folder may contain the repository one level down |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
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.