kanso-commit

kanso-commit is a skill for Claude Code, Codex from blakecyze/kanso. It costs 31 tokens per session (1,889 once invoked), scanned A, original, MIT.

A guide for splitting code changes into small, focused Git commits, with messages that explain the reason for each change. Git is the version-control system used to track and share code history.

In plain words
What is it for?
Use it when staging work, choosing commit boundaries, and writing commit messages that match the repository's existing style.
Why use it?
It prevents unrelated edits from being bundled together and helps commits stay easier to review, undo, or move to another branch.

Skill for Claude CodeCodex

Part of the kanso plugin — 9 skills, 1 hook shipped together

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/blakecyze/kanso/kanso-commit
Any agent
npx skills add blakecyze/kanso --skill kanso-commit
Clone the repo
git clone --depth 1 https://github.com/blakecyze/kanso

Made for: Claude Code, Codex.

Or install kanso, the plugin that ships this one along with the rest of its 9 skills, 1 hook.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for kanso-commit

README.md
[![agentmods](https://agentmods.dev/badge/skills/blakecyze/kanso/kanso-commit.svg)](https://agentmods.dev/skills/blakecyze/kanso/kanso-commit)
Your own site
<a href="https://agentmods.dev/skills/blakecyze/kanso/kanso-commit"><img src="https://agentmods.dev/badge/skills/blakecyze/kanso/kanso-commit.svg" alt="Measured on agentmods" height="20"></a>
Per session 31 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,889 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 $0.00031 $0.01889
Opus 5 $0.00015 $0.00945
Sonnet 5 $0.00006 $0.00378
Haiku 4.5 $0.00003 $0.00189

Measured 3d ago against content hash 2557f948fefa, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

kanso-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 3d 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.

skills/kanso-commit/SKILL.md · 177 lines

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.

kanso-commit

Splits the working tree into logically atomic commits and writes messages that answer why rather than what. Detects and matches the repo's existing convention.

The principles from kanso-principles apply. Messages earn their place the same way code does.

The atomic unit

An atomic commit is one logical change. Not one line, not one file — one thought. The test: can the commit be described in a single short imperative sentence without using "and"?

Each atomic commit must be:

  • Independently revertable. Reverting it doesn't cause collateral damage.
  • Independently cherry-pickable. It can move to another branch cleanly.
  • Stable. Tests pass (or at least don't newly fail) at this commit.
  • Single-concern. Refactor and feature are separate commits, always. Formatting and logic are separate commits.

Over-atomising is a real failure mode. A three-line bug fix does not need three commits.

Before doing anything

  1. Check the status. !git status --short and !git diff --stat to see what's staged and unstaged.
  2. Detect the convention. !git log -n 30 --oneline and read 3-5 recent commits in full. The repo's existing style wins over any default.
  3. Detect the tooling. Look for commitlint.config.*, .commitlintrc*, cz-config*, package.json scripts mentioning commit. These hard-constrain the format.

Convention detection

The repo's commit history dictates the format. In priority order:

  1. commitlint config present → use whatever it enforces. No exceptions.
  2. Conventional Commits pattern in recent history (feat:, fix:, chore:) → match it.
  3. Custom scoped pattern ([Component] subject, subsystem: subject, Fixed #NNNNN --) → match it exactly. Don't modernise.
  4. Plain imperative subjects with no prefix → match that. Don't introduce prefixes.
  5. No clear pattern, fewer than 10 commits in history → use Conventional Commits as a sensible default and tell the user.

Don't impose a new convention onto an existing repo. If the history is messy, match the least-bad recent pattern rather than inventing a new one.

Read the full file on GitHub · 177 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. 3d ago First seen · 177 lines · 31 tokens per session scan A 2557f948fefa

Subscribe to this mod's changes

kanso-commit is a skill published in the GitHub repository blakecyze/kanso (3 stars, last pushed 1mo ago), licensed MIT. It adds 31 tokens to every session and 1,889 once invoked, about $0.0002 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.