# Create Claude Code Output Styles

Claude Code output styles are easy to misconfigure. This skill guides valid style files, placement, activation, and testing.

## Install

```bash
npx skillstore add emz1998/generating-output-styles
```

## Metadata

- Status: approved
- Slug: emz1998-generating-output-styles
- Skillstore revision: r1
- Version status: missing
- Tree hash: 1a86c17060766790f4475558413c26188296d930f05a36d251fd5b6f37dde669
- Author: Emz1998
- GitHub username: Emz1998
- License: MIT
- Repository: https://github.com/Emz1998/avaris-ai/tree/master/.claude/skills/output-style
- Ref: 30c73eac2afe762f6aa9c4553158769369d47351
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: filesystem, external\_commands
- Quality score: 74
- Public page: https://skillstore.pages.dev/skills/emz1998-generating-output-styles
- Manifest: https://skillstore.pages.dev/api/skills/emz1998-generating-output-styles/manifest

## Capabilities

- Creates Markdown output style files with required frontmatter fields.
- Explains when to keep or remove Claude Code coding instructions.
- Guides user-level and project-level output style placement.
- Provides naming, description, and formatting rules for style files.
- Includes activation and testing steps with the output-style menu.
- Uses the included output style reference documentation.

## Use Cases

- Team Response Standards: Create a shared project style that keeps responses consistent for a repository.
- Learning Mode Customization: Build a style that adds explanations while retaining coding instructions.
- Documentation-Focused Work: Create a style that emphasizes docs, trade-offs, and structured implementation notes.

## Prompt Templates

### Create a Simple Style

```
Create a Claude Code output style named Learning Guide. It should explain decisions clearly and keep coding instructions enabled.
```

### Convert Team Preferences

```
Turn these team response preferences into a valid Claude Code output style. Keep coding instructions enabled and include a user-friendly description: [preferences].
```

### Design a Domain Style

```
Create a project-level output style for security review sessions. Preserve security boundaries, keep coding instructions enabled, and include testing steps.
```

### Audit an Existing Style

```
Review this output style for valid frontmatter, clear behavior, safe instructions, and activation readiness. Suggest precise improvements: [style text].
```

## Limitations

- Does not install or activate styles without file changes in Claude Code configuration.
- Cannot override Claude Code tool permissions or security restrictions.
- Generated styles still need review for organization policies and safety.
- Behavior depends on active Claude Code output style support.

## Best Practices

- Review generated instructions before saving them into user-level configuration.
- Use keep-coding-instructions when the style supports software engineering work.
- Test each style with representative tasks before sharing it with a team.

## Anti Patterns

- Do not use output styles to bypass tool permissions or security rules.
- Do not save broad behavior changes at user level when only one project needs them.
- Do not create vague style instructions that conflict with Claude Code workflows.

## Security Audit

- Audited at: 2026-07-05T15:11:03.203\+00:00
- Summary: All Ruby or shell execution findings are Markdown formatting examples, not executable commands. The hidden-directory references point to documented Claude Code output style locations and are user-visible configuration guidance. No prompt injection attempt, exfiltration intent, or system reconnaissance behavior was found.

## Stats

- Views: 0
- Downloads: 4
- Favorites: 0
- Popularity score: 0