commit

A guided process for preparing a Git commit, which is a saved set of code changes. It checks the staged changes and proposes a commit message and files.

In plain words
What is it for?
Use it to review staged or unstaged changes, check for security issues, decide what belongs in the commit, and prepare the exact commit proposal.
Why use it?
It helps catch secrets, credentials, environment files, and unfinished work before anything is committed. The commit still waits for your explicit approval.

Skill for Claude CodeCodexCursor

Install

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.

agentmods
npx agentmods add skills/hlalljie/agent-workflow-presets/commit
Any agent
npx skills add hlalljie/agent-workflow-presets --skill commit
Clone the repo
git clone --depth 1 https://github.com/hlalljie/agent-workflow-presets

Made for: Claude Code, Codex, Cursor.

Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,134 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What 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.

ModelPer sessionOnce invoked
Fable 5 $0.00029 $0.01134
Opus 5 $0.00015 $0.00567
Sonnet 5 $0.00006 $0.00227
Haiku 4.5 $0.00003 $0.00113

Measured yesterday against content hash 76205f88016a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 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.

.cursor/skills/commit/SKILL.md · 70 lines

How it starts

The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Commit

Process

  1. Run git diff --staged and git status (if nothing staged, note unstaged changes).
  2. Security review — no secrets, credentials, or env files staged; sensitive paths ignored.
  3. Loose ends — only if something needs a follow-up. If there are none, omit this (do not write "none", do not explain what you are not doing).
  4. Propose — commit message, files to stage, Ready to commit (y/n):, stop. Do not run git add or git commit.
  5. Wait for explicit approval.
  6. Only then run git add and git commit.

Starting vs approving

  • /commit, "prepare a commit", "go ahead and /commit", "run the commit skill" mean: run steps 1–4 (including the full proposed commit message and files to stage), print Ready to commit (y/n):, then stop. That is not permission to run git commit yet.
  • Approval is a separate message after that proposal: e.g. y, "yes", "approved", or a short edit ("use that message" / tweak subject only). Never treat "go ahead" in the same message as /commit as approval to commit unseen.
  • The user must be able to read the proposed subject and body before approving.

Proposal output: short and scannable. If this commit finishes the last open task for a plan phase: state that clearly in the proposal to the user and in the commit message (subject or bullet); include updating that phase's ### line in plan.md with — complete (same commit as task checkboxes — see the "Phases" section at the top of that plan). If it does not complete a phase, say nothing about phases — do not list non-events.

This applies no matter how the user phrases it — "commit this", "create a commit", "prepare a commit", "/commit", "go ahead and /commit" all mean propose first (steps 1–4), commit only after a clear follow-up approval (step 6).

Commit message format

Subject: State what changed for the user or product (outcome / intent), not a dump of file moves. Someone reading only the subject should understand the change without opening the diff.

Read the full file on GitHub · 70 lines

Changes

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.

  1. yesterday First seen · 70 lines · 29 tokens per session scan A 76205f88016a

Subscribe to this mod's changes

commit is a skill published in the GitHub repository hlalljie/agent-workflow-presets (2 stars, last pushed 1mo ago), licensed MIT. It adds 29 tokens to every session and 1,134 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.