agent-creator
Use when creating expert agents. Generates agent.md with frontmatter, hooks, required sections, and skill references.
Detects optimal commit type from git changes. Use when analyzing commits, determining commit type, or before committing.
$ npx -y skills add fusengine/agents --skill commit-detection --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/commit-detectionContext preview
The summary Claude sees to decide when to auto-load this skill.
Detects optimal commit type from git changes. Use when analyzing commits, determining commit type, or before committing.
name: commit-detection description: Detects optimal commit type from git changes. Use when analyzing commits, determining commit type, or before committing. allowed-tools: Bash, Read, Grep
<objective> Detects the optimal conventional commit type (feat/fix/docs/style/refactor/perf/test/build/ci/chore) from staged and unstaged git changes, using file-pattern categorization (docs/test/config/ci/build/src) plus diff-keyword heuristics ("fix", "bug", "optimize", etc.), and derives a scope suggestion from the primary changed directory (e.g. `src/api/auth.ts` → `auth`). Covers the full detection algorithm, the rule cascade, scope extraction, and worked examples. Does not perform the commit itself — routes to the matching `/commit-pro:<type>` command, or to the `commit` skill for full smart analysis when no rule matches cleanly. </objective>
Expert knowledge for detecting the optimal conventional commit type.
# Get modified files git diff --name-only git diff --staged --name-only # Get change statistics git diff --stat git diff --staged --stat # Check for keywords in diff git diff | grep -i "fix\|bug\|error" | head -5
| Category | File Patterns | |----------|---------------| | docs | `*.md`, `*.txt`, `*.rst`, `README*`, `CHANGELOG*` | | test | `*.test.*`, `*.spec.*`, `__tests__/*`, `test/*` | | config | `*.json`, `*.yml`, `*.yaml`, `*.toml`, `.*rc` | | ci | `.github/*`, `.gitlab-ci.yml`, `Jenkinsfile` | | build | `package.json`, `Makefile`, `webpack.*`, `vite.*` | | style | Only whitespace, formatting changes | | src | `*.ts`, `*.js`, `*.py`, `*.go`, `*.rs`, etc. |
IF only docs files changed:
→ docs
IF only test files changed:
→ test
IF only config/build files changed:
→ chore
IF only CI files changed:
→ ci
IF diff contains "fix", "bug", "error", "issue", "resolve":
→ fix
IF new files added with business logic:
→ feat
IF files renamed/moved without logic change:
→ refactor
IF performance keywords ("optimize", "perf", "speed", "cache"):
→ perf
IF formatting only (whitespace, semicolons):
→ style
DEFAULT:
→ Use /commit-pro:commit for smart analysisExtract scope from primary directory:
src/components/Button.tsx → ui or button src/api/auth.ts → auth lib/utils/date.ts → utils server/routes/user.ts → user
| Type | When | Version Bump | |------|------|-------------| | `feat` | New functionality | PATCH | | `fix` | Bug correction | PATCH | | `docs` | Documentation only | PATCH | | `style` | Formatting only | PATCH | | `refactor` | Code restructure | PATCH | | `perf` | Performance | PATCH | | `test` | Tests only | PATCH | | `build` | Build/deps | PATCH | | `ci` | CI/CD config | PATCH | | `chore` | Maintenance | PATCH |
> MINOR/MAJOR bumps are **manual user decisions**, never automatic.
See the `post-commit` skill for universal CHANGELOG, version bump, and tag logic (works for all repos).
**Example 1: Only README changed**
Files: README.md → /commit-pro:docs
**Example 2: New component + test**
Files: src/Button.tsx, src/Button.test.tsx → /commit-pro:feat (primary is new feature)
**Example 3: Fix in existing file**
Files: src/api/auth.ts Diff contains: "fix login bug" → /commit-pro:fix
A plugin ecosystem that turns Claude Code into a supervised, multi-agent development environment.
Repo: fusengine/agents
Use when creating expert agents. Generates agent.md with frontmatter, hooks, required sections, and skill references.
Use when starting ANY development task -- feature, bug fix, refactor, hotfix (triggers: implement, create, build, fix, add feature, refactor, develop).
Use when creating a feature/component or adding functionality. Fires BEFORE APEX Analyze to refine requirements via structured questioning.
Use before a root-cause, done/verified claim, irreversible action, or 2nd-time fix reaches the owner (APEX or plain conversation); also fires at every…
Use when validating code quality after modifications -- SOLID compliance, DRY duplication, linter errors, architecture violations. Do NOT use for functional…
Use when an expert agent self-reviews and self-corrects code after the Execute phase, before sniper validation (BMAD-METHOD elicitation techniques).