AI Skills
Pre-defined skills for AI assistants and LLM agents to help with common tasks.
Overview
ZeroStarter includes a .skills directory containing structured instructions for AI assistants and LLM agents. These skills help AI tools like Cursor, GitHub Copilot, and other AI coding assistants perform common tasks correctly and consistently.
Skills are written in Markdown with frontmatter metadata, making them both human-readable and machine-parseable.
Available Skills
Git Commit
Location: .skills/git/commit/SKILL.md
Creates atomic commits with conventional commit message format.
Triggers: When user asks to "commit", "save changes", or "create a commit"
Workflow:
- Inspect Changes: Review both staged and unstaged changes
- Stage Changes: Stage related changes together (one logical unit per commit)
- Commit: Create commit with conventional message format
- Push (Optional): Only push if user explicitly requests it
Default Behavior: The skill stages and commits changes but does not push automatically. This prevents accidental pushes to remote branches and gives users control over when changes are shared.
Push Confirmation: To push after committing, the user must explicitly request it (e.g., "commit and push", "push the changes", or respond affirmatively when asked). If the user only asks to "commit" or "save changes", the skill stops after creating the commit.
Message Format:
<type>(<scope>): <subject>
<body>| Element | Description |
|---|---|
| Type | feat, fix, refactor, docs, chore, test, perf, style, ci |
| Scope | Optional area in parentheses (package, component, feature name) |
| Subject | Imperative mood, lowercase, no period, ≤80 chars |
| Body | Optional; explain "why" not "what"; wrap at 80 chars |
Examples:
feat(auth): add OAuth provider support
fix(api): prevent duplicate webhook delivery
refactor(web): extract auth middleware into separate module
docs(readme): update installation steps
chore(deps): bump dependencies to latest versionsTurbo Build Graph
Location: .skills/turbo/generate-build-graph/SKILL.md
Generates a styled SVG visualization of Turborepo task dependencies.
Triggers: When user asks to "visualize dependencies", "generate build graph", or "update pipeline diagram"
Workflow:
- Generate Raw Graph:
npx turbo run build --graph=graph.svg - Clean Up Labels: Remove prefixes and internal placeholders
- Style for Dark Mode: Apply transparent background and GitHub blue theme
- Move to Assets: Save to
.github/assets/graph-build.svg
Output: A dark-mode-friendly SVG that can be embedded in README:
Skill Structure
Each skill follows a standard structure:
.skills/
└── <category>/
└── <skill-name>/
└── SKILL.mdSKILL.md Format
---
name: skill-name
description: Brief description of what the skill does and when to use it.
---
# Skill Title
Description of the skill.
## Workflow
### 1. First Step
```bash
command to run
```- Notes about this step
- Important considerations
2. Second Step
...
## Creating Custom Skills
To create a new skill:
1. Create a directory: `.skills/<category>/<skill-name>/`
2. Add a `SKILL.md` file with:
- Frontmatter with `name` and `description`
- Clear workflow steps
- Code examples with commands
- Notes and edge cases
**Example Custom Skill**:
```markdown
---
name: db-migration
description: Creates and runs database migrations. Use when user asks to update database schema.
---
# Database Migration
Creates type-safe database migrations with Drizzle.
## Workflow
### 1. Create Migration
```bash
bun run db:generate2. Review Migration
Check the generated SQL in packages/db/drizzle/
3. Apply Migration
bun run db:migrate
## Integration with AI Tools
### Cursor
Skills in `.skills/` are automatically available to Cursor's AI assistant when referenced in `AGENTS.md`:
```markdown
## Skills
There are few skills that are used in the project to help the user with the tasks. You can find them in the `<root>/.skills` directory.GitHub Copilot
Copilot can reference skills when you mention them in comments or prompts.
Custom AI Assistants
The structured format makes skills easy to parse programmatically for custom AI integrations.
Best Practices
- Keep skills focused: One task per skill
- Include examples: Show expected inputs and outputs
- Document edge cases: Handle errors and special situations
- Use standard commands: Prefer project scripts over raw commands
- Version control: Skills evolve with the project