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/zarl-prog/smart-git-commit/smart-git-commitnpx skills add Zarl-prog/Smart-git-commit --skill smart-git-commitgit clone --depth 1 https://github.com/Zarl-prog/Smart-git-commitWhat 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 | $0.00148 | $0.01888 |
| Opus 5 | $0.00074 | $0.00944 |
| Sonnet 5 | $0.00030 | $0.00378 |
| Haiku 4.5 | $0.00015 | $0.00189 |
Grade A, and why
smart-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 — 277 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Smart Git Commit Skill
Produces gold-standard Git commits: tested, atomic, secure, documented, and traceable. Follow every phase below in order. Never skip phases unless the user explicitly says so.
Phase 0 — Read Project Rules
Check for CLAUDE.md, .gitmessage, .git/COMMIT_TEMPLATE at repo root. If found, those rules override this skill's defaults.
cat CLAUDE.md 2>/dev/null && echo "→ Found CLAUDE.md"
cat .gitmessage 2>/dev/null && echo "→ Found .gitmessage"
Show what rules were loaded before proceeding.
Phase 1 — Diff Analysis
Run full diff analysis before touching git add:
git status
git diff --stat # which files, how many lines
git diff # full diff for small changesets
git log --oneline -5 # recent commit context
Categorize every changed file:
| Path pattern | Category |
|---|---|
src/ lib/ app/ |
feature/fix/refactor |
tests/ *.test.* |
test |
docs/ *.md |
docs |
package.json deps |
chore |
.github/ Makefile |
tooling |
If files span more than one category → flag for atomic split in Phase 4.
Read references/atomic-patterns.md if split is needed.
Phase 2 — Security Scan (Never Skip)
Run the automated security scanner:
bash scripts/scan-secrets.sh
- Exit 0 → show "✓ Clean" and continue
- Exit 1 → HARD STOP. Show findings. Do not proceed until clean.
If secrets found:
git reset HEAD <file>to unstage- Replace with env var or placeholder
- Add to
.gitignoreif needed - Rotate exposed credentials if already pushed
Read references/security-rules.md for the full pattern list.
Phase 3 — Test Gate (Never Skip)
Auto-detect and run the project's test suite:
bash scripts/detect-test-runner.sh
- status = "pass" → show test count and continue
- status = "fail" → HARD STOP. Fix failures first, then commit fix + feature together.
- status = "not_found" → warn user, ask if they want to proceed anyway
What ships with it
28 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- contributor/CONTRIBUTOR-SKILL.md 5.6 KB
- contributor/references/first-contribution.md 8.1 KB
- contributor/references/pr-anatomy.md 9.3 KB
- contributor/references/review-etiquette.md 6.6 KB
- contributor/scripts/branch-name.sh 4.7 KB runs code
- contributor/scripts/fork-check.sh 3.2 KB runs code
- contributor/scripts/pr-readiness.sh 6.8 KB runs code
- contributor/scripts/review-response.sh 5.1 KB runs code
- contributor/templates/pr-body-full.md 959 B
- contributor/templates/pr-title.md 2.7 KB
- contributor/templates/review-response.md 3.7 KB
- references/atomic-patterns.md 5.3 KB
- references/commit-types.md 4.9 KB
- references/message-examples.md 9.7 KB
- references/release-workflow.md 5.2 KB
- references/security-rules.md 5.6 KB
- scripts/create-pr.sh 4.7 KB runs code
- scripts/detect-test-runner.sh 3.8 KB runs code
- scripts/generate-changelog.sh 4.7 KB runs code
- scripts/scan-secrets.sh 3.9 KB runs code
- scripts/split-commits.sh 6.2 KB runs code
- templates/.gitmessage 572 B
- templates/CLAUDE.md.example 1.7 KB
- templates/pr-body.md 618 B
- tests/fixtures/mixed-diff.txt 4.3 KB
- tests/fixtures/secret-diff.txt 1.2 KB
- tests/README.md 1.8 KB
- tests/test-scenarios.md 4.7 KB
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 · 277 lines · 148 tokens per session scan A c984b24efac1
smart-git-commit is a skill published in the GitHub repository Zarl-prog/Smart-git-commit (6 stars, last pushed 21d ago), licensed MIT. It adds 148 tokens to every session and 1,888 once invoked, about $0.0007 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
appium-python-expert
Specialist skill for mobile E2E testing with Appium 2.x + Python (pytest) for Android and iOS. Use this skill whenever the user asks about: setting up Appium with Python, writing mobile test cases, configuring Android/iOS drivers (UIAutomator2, XCUITest), implementing Page Object Model for mobile, running tests with…
selenium-cucumber-expert
Use when writing E2E tests with Selenium WebDriver and Cucumber BDD in Java, setting up BDD test infrastructure with Maven or Gradle, writing Gherkin feature files, implementing step definitions in Java, configuring Page Object Model or Screenplay Pattern, setting up parallel execution, generating ExtentReports or…
diffler
Generate and deliver a graded comprehension quiz from the current Git branch diff, either in the local terminal or through Google Forms. Use when a user asks to prove, test, or verify understanding of branch changes, a feature branch, or a pull request.
playwright-cucumber-expert
Use when writing E2E tests with Cucumber BDD and Playwright, setting up BDD test infrastructure, writing Gherkin feature files, implementing step definitions, configuring hooks and World, or integrating Cucumber with CI/CD. Invoke for Cucumber, BDD, Gherkin, feature files, step definitions, Given When Then, hooks…
playwright-automation-expert
Use when writing E2E tests with Playwright, setting up test infrastructure, debugging flaky browser tests, organizing project structure, or testing REST APIs. Invoke for browser automation, E2E tests, Page Object Model, test flakiness, visual testing, project scaffolding, folder layout, API testing, JSON schema…
karpathy-guidelines
Tool-agnostic behavioral guidelines for AI coding assistants. Use when writing, reviewing, debugging, or refactoring code to reduce overengineering, surface ambiguity, make surgical changes, and define verifiable success criteria.