Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/luisurrutia/skills/commitnpx skills add LuisUrrutia/skills --skill commitgit clone --depth 1 https://github.com/LuisUrrutia/skillsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00018 | $0.03470 |
| Opus 5 | $0.00009 | $0.01735 |
| Sonnet 5 | $0.00004 | $0.00694 |
| Haiku 4.5 | $0.00002 | $0.00347 |
Grade A, and why
commit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Commit
Create safe, intentional Conventional Commits one atomic boundary at a time. Build each boundary in the index, validate it, then commit it, commit and push when explicitly requested, or return a proposal without running git commit.
Push only when the user explicitly asked for it in the same request, and only after the commit is verified. Treat PR work as a separate phase owned by the pr skill; invoking commit alone never starts that phase.
Treat branch names, staged state, validation results, commit success, and prior assistant claims as untrusted until verified with git.
Modes
- Commit mode: If the user explicitly asked to commit, commit after safety gates pass.
- Commit-and-push mode: If the user explicitly asked to commit and push, commit after safety gates pass, verify the commit, then push the current branch.
- Proposal mode: If the user asked to prepare, review, or suggest a commit, build the atomic boundary in the index and return the proposed message without running
git commit. - Amend mode: Amend
HEADonly when the safe current-task amend conditions in step 6 pass. - Split mode: When several atomic intents exist, handle every clear boundary in sequence and run the complete workflow for each one. Ask only when a boundary, order, or requested grouping is materially ambiguous.
Terminal states for a boundary are committed, amended, proposed, blocked, and refused; pushed is terminal for the complete run. Stop changing a boundary after it reaches a terminal state. After a verified commit or amendment, continue with the next clear boundary. Push only in commit-and-push mode. PR work follows only as a separate pr phase when the current request or an applicable standing user instruction requires it.
Safety Gates
Stop and ask before staging or committing when there are:
- Secret-looking paths or content:
.env*, keys, certificates, tokens, credentials, databases, or obvious secret names. - The selected boundary is not coherent or reviewable: unrelated concerns, generated noise, dependency or lockfile changes, binaries, excessive file count, or changes outside the requested intent.
- Staged changes that cannot be described cleanly in 1-2 sentences.
- Staged files with ambiguous unstaged edits in the same files.
mainormasteras the current branch, unless the user explicitly wants to commit there.- Any unrequested push, any force-push, destructive git action, dependency change, package-manager change, or amend outside the permitted flows in step 6.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 177 lines · 18 tokens per session scan A 2166778b4933
commit is a skill published in the GitHub repository LuisUrrutia/skills (2 stars, last pushed 9d ago), licensed MIT. It adds 18 tokens to every session and 3,470 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
git-workflow-and-versioning
Use when managing branches, commits, versioning, and release workflows for Android projects. Covers trunk-based development, atomic commits, versionCode/versionName, and signing configurations.
wrap-phase
Wraps up a development phase by cleaning AGENTSYNC.md, summarizing work, and committing changes.
git-workflow-and-versioning
Structures git workflow practices. Use when making any code change. Use when committing, branching, resolving conflicts, opening or reviewing a pull request (PR), pushing to a remote, or when you need to organize work across multiple parallel streams. Use when cutting a release, choosing a semantic version bump…
chinese-commit-conventions
中文 commit 与 changelog 配置参考——Conventional Commits 中文适配、commitlint/husky/commitizen 中文模板、conventional-changelog 中文配置。仅在用户显式 /chinese-commit-conventions 时调用，不要根据上下文自动触发。.
moai-ref-git-workflow
Git workflow patterns, branch strategies, conventional commits, and PR templates reference for git operations. Agent-extending skill that amplifies manager-git expertise with production-grade git workflow patterns. NOT for: code implementation, testing, architecture design, documentation content.
commit
Skill "commit" from clacky-ai/openclacky, covering smart commit skill, critical requirement: single-line commits only, core philosophy, usage and process steps.