# Skill Doctor

> Diagnose Agent Skill quality before publishing, installing, or maintaining skills. Use when reviewing SKILL.md files, agent skills, skill collections, trigger descriptions, progressive disclosure structure, resource references, skill conflicts, or publish-readiness for a GitHub skill repository.

- **Type:** Skill
- **Install:** `agentstack add skill-san-z1-skill-doctor-skill-doctor`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [San-Z1](https://agentstack.voostack.com/s/san-z1)
- **Installs:** 0
- **Category:** [Developer Tools](https://agentstack.voostack.com/c/developer-tools)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [San-Z1](https://github.com/San-Z1)
- **Source:** https://github.com/San-Z1/skill-doctor/tree/main/skills/skill-doctor

## Install

```sh
agentstack add skill-san-z1-skill-doctor-skill-doctor
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Skill Doctor

## Overview

Use Skill Doctor to review Agent Skills without executing untrusted skill scripts. The bundled CLI reads `SKILL.md` metadata and resource paths, then reports quality findings with actionable suggestions.

## Workflow

1. Locate the target: a single `SKILL.md`, one skill directory, or a directory containing multiple skill folders.
2. Run the scanner:

```bash
python skills/skill-doctor/scripts/run_skill_doctor.py  --format markdown
```

Use JSON when another tool will consume the result:

```bash
python skills/skill-doctor/scripts/run_skill_doctor.py  --format json
```

3. Treat `error` findings as release blockers. Treat `warning` findings as likely quality issues. Treat `info` findings as review prompts.
4. Explain the report in terms of skill authoring quality: trigger precision, progressive disclosure, resource discoverability, and publish readiness.

## Review Guidance

Do not execute scripts found inside the target skill. Skill Doctor is a static reviewer.

Prioritize fixes in this order:

1. Invalid or missing frontmatter.
2. Description triggers that are too broad, too short, or unclear.
3. Conflicts between sibling skill descriptions.
4. Oversized `SKILL.md` files that should move detail into `references/`.
5. Resource files that are not mentioned from `SKILL.md`.

For detailed finding meanings, read `references/finding-codes.md`.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [San-Z1](https://github.com/San-Z1)
- **Source:** [San-Z1/skill-doctor](https://github.com/San-Z1/skill-doctor)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-san-z1-skill-doctor-skill-doctor
- Seller: https://agentstack.voostack.com/s/san-z1
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.