Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Git Conventions

BSecurity

Use when committing, squashing, merging, rebasing, or preparing changes for a pull request

4 stars
0 votes
0 copies
0 views
Added 9/23/2026
ai-agentsgobashterraformgitapi

Works with

claude codeapi

Security Analysis

B85/100
highPerforms destructive filesystem operations

Pro shows the line behind each finding and how to fix it

Scanned 9/23/2026

$npx -y skills add metraton/gaia --skill git-conventions --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Git Conventions?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Git Conventions
[![Security: B — Skills Directory](https://www.skillsdirectory.com/api/skills/metraton-git-conventions/badge)](https://www.skillsdirectory.com/skills/metraton-git-conventions)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: git-conventions
description: Use when committing, squashing, merging, rebasing, or preparing changes for a pull request
---

# Git Conventions

## Commit Format

| Element | Rule |
|---------|------|
| Format | `type(scope): short description` |
| Types | feat, fix, refactor, docs, test, chore, ci, perf, style, build |
| Scope | Optional, reflects module/area changed |
| Subject | Max 72 chars, lowercase start, imperative mood, no period, no emoji |
| Body | Optional, blank line after subject, 72 char line wrap |
| Footers | `BREAKING CHANGE:`, `Refs:`, `Closes:`, `Fixes:`, `Implements:`, `See:` |

## Examples

```
feat(helmrelease): add Phase 3.3 services
fix(pg-non-prod): correct API key environment variable mappings
refactor: simplify context provider logic
chore(deps): update terraform to v1.6.0
```

## Multi-Line Body (No Heredoc, No Temp File)

Repeat `-m` once per paragraph -- `git -C /abs commit -m "type: subject" -m
"First paragraph." -m "Second paragraph."` -- no heredoc, no temp file. For
exact wraps, use `git commit -m "$(cat <<'EOF' ... EOF)"` instead
(`command-execution` Rule 7). Never attach a bare, unquoted heredoc
directly to git (`git commit -F - <<EOF ... EOF`) -- measured, the
classifier misreads a prose line as a command and blocks with a spurious
T3 (`Verb: 'rm' (MUTATIVE)` on text that only *mentions* `rm -rf`).

## Git Path Flags

Target the repo with `git -C /absolute/path <verb>` -- this is the canonical
form in a subagent. The T3 consent gate parses `-C` as a flag and classifies
the command identically to the bare verb (`git -C /repo push` is T3 `push`,
same as `git push`); there is NO bypass of Gaia's gate. The full discipline --
why one byte-identical form prevents the approval loop, and why a separate `cd`
call cannot work (the cwd resets between Bash calls) -- lives in
`command-execution` Rule 7; follow it there.

Two narrow caveats:
- Prefer `-C /abs` (short flag) or `--git-dir=/abs` (equals form); both are
  cleanly absorbed. Avoid the SPACE-separated `--git-dir /abs` / `--work-tree
  /abs` forms -- the path leaks into the first non-flag token and shifts the
  subcommand the local-safe guard reads (a mutative verb like `push` is still
  caught by the fallback scanner, but commit-message-leak protection is lost).
- The historical warning that a leading path flag "bypasses all rules silently"
  described Claude Code's NATIVE settings.json prefix matching (`Bash(git
  commit:*)`), which Gaia used in 3.2.1 but no longer ships -- enforcement is
  now the PreToolUse hook, which is `-C`-robust. If a user adds their own
  `Bash(git ...:*)` allow rules, a leading path flag can still shift that
  prefix, but that affects only their auto-allow convenience, never Gaia's
  consent gate.

## Push Defaults

Push to the feature branch. Only push directly to `main` when explicitly
instructed or when the work is already on main. Force-push (`--force`)
requires explicit user instruction.

## Hook Enforcement

The `commit_validator.py` hook validates against standards inlined as
module-level constants in that file (`TYPE_ALLOWED`, `SUBJECT_MAX_LENGTH`,
`SUBJECT_RULES`, `BODY_MAX_LINE_LENGTH`) -- it covers the conventional-commits
format, subject, and body rules. Forbidden-footer detection lives separately
in `bash_validator` (hardcoded there). Format violations block the commit.
Body line length triggers warnings only.

## Pull Request Body

A PR is written retrospectively, after the commits already exist -- it
explains the change to a reviewer who was not present while it was made, in
English, self-contained, with no reference to the process that produced it
(no task IDs, no plan/sprint labels, no "as requested"). The title reuses the
same `type(scope): outcome` rule as the commit subject above -- no plan
jargon there either.

| Section | Content |
|---------|---------|
| Context / Situation | The state that existed before this change and why it needed one |
| What changed | The concrete change made, stated plainly |
| Impact / what it solves | The outcome the change produces for the system or the user |
| Plan impact | For IaC changes: the plan summary -- resources to add, change, destroy |
| Verification | How the change was verified -- tests run, dry-run/plan output, manual check |
| Risk / rollback (optional) | What could go wrong and how to revert, when the change carries real risk |
| Links (optional) | Related PRs, issues, or docs -- only when they add context a reviewer needs |

This structure is advisory, not enforced: `commit_validator.py` validates
only the commit message format above -- it does not parse or gate the PR
body. Writing a body without this structure will not block anything; the
structure exists so a reviewer gets the same shape every time.

Attribution

metratonmetraton
View sourceSee grades on GitHubMore from metraton →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →