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/gabrieldabbah/genesis/git-commitnpx skills add gabrieldabbah/genesis --skill git-commitgit clone --depth 1 https://github.com/gabrieldabbah/genesisWhat 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 | $0.00094 | $0.01388 |
| Opus 5 | $0.00047 | $0.00694 |
| Sonnet 5 | $0.00019 | $0.00278 |
| Haiku 4.5 | $0.00009 | $0.00139 |
Grade A, and why
git-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 yesterday.
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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Commit and push
The message comes from the diff, not from the conversation. What you remember doing and what is actually staged diverge constantly — a message written from memory describes a commit that was not made.
Portability
Every command here is POSIX shell and runs on macOS, Linux, WSL2 and Git Bash. On native Windows without
Git for Windows, Claude Code runs commands through PowerShell, which
cannot execute these — translate them or install Git for Windows. No GNU-only flags are used, so BSD
userland on macOS behaves identically. git is the only required binary.
1. Read the state before touching it
git rev-parse --abbrev-ref HEAD # which branch
git status --porcelain # what is modified, staged, untracked
git diff --staged # if something is already staged, this is the commit
git diff # otherwise this is the candidate
Someone else's staged work is not yours to commit. A non-empty index you did not create belongs to
another session or an interrupted one. Commit only what this task changed — stage those paths by name.
git add -A sweeps a working tree you have not read.
2. Branch
Commit on the current branch. Read the repository's own branch policy — CLAUDE.md, CONTRIBUTING.md, the
default branch on the remote — and follow it; where it names a working branch, that is where commits go.
- On the default branch (
main/master) in a repository that has a separate working branch: ask before committing, unless the operator asked for a commit there specifically. - Creating a branch is outside this skill. A request to commit is not a request to branch.
3. Group before you write
One logical change per commit. Where the working tree holds two unrelated changes, that is two commits with two messages, staged by path — not one commit with a message that has to say "and".
4. Message
<type>[optional scope]: <description>
[optional body — why, not what; the diff already says what]
[optional footer]
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.
- yesterday First seen · 140 lines · 94 tokens per session scan A c6136957282c
git-commit is a skill published in the GitHub repository gabrieldabbah/genesis (4 stars, last pushed 28d ago), licensed MIT. It adds 94 tokens to every session and 1,388 once invoked, about $0.0005 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
watch
File sentinel that monitors the working directory for changes and marker comments, then auto-triggers appropriate skills. Poll-based via git diff against the last scan commit. Writes intake items for batch processing and routes marker actions through /do. Use for automatic reactions to file changes; do NOT use for…
pr-watch
Local PR watcher. Monitors CI status, automatically fixes failing checks by reading failure logs and applying targeted fixes, then optionally merges when all checks pass. Local CLI analog to Claude Code's cloud auto-fix feature.
qa
Browser-based QA verification. Launches a real browser, navigates the app, clicks buttons, fills forms, and tests user flows. Works as a standalone skill or as a phase end condition in campaigns. Requires Playwright (optional dependency, graceful skip if not installed).
review
5-pass structured code review — correctness, security, performance, readability, consistency.
live-preview
Mid-build visual verification loop. Takes screenshots of components during construction, not just after. Catches visual regressions and invisible features before they compound. Requires Playwright or similar screenshot tool.
wiki
Markdown-first knowledge base where the LLM acts as librarian. Ingests raw sources, compiles and interlinks topic files, self-maintains an index. No vector DB or embeddings required -- uses LLM-native navigation over structured markdown up to 400K words.