# Validate Agent Skill Files

Skill files can drift from the Agent Skills specification and trigger unreliably. This skill validates structure, frontmatter, descriptions, and evaluation examples.

## Install

```bash
npx skillstore add supercent-io/skill-standardization
```

## Metadata

- Status: approved
- Slug: supercent-io-skill-standardization
- Version: 2.0
- Author version: 2.0
- Skillstore revision: r1
- Version status: invalid
- Tree hash: 2187f02828a4967cab9533c81700fbfe9d900360ffa715f5f8e8f43ebb0569a5
- Author: supercent-io
- GitHub username: supercent-io
- License: MIT
- Repository: https://github.com/supercent-io/skills-template/tree/main/.agent-skills/skill-standardization/
- Ref: f93e9bb0daca99badb6a7e574b97737155d57cb3
- 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: external\_commands, network
- Quality score: 76
- Public page: https://skillstore.pages.dev/skills/supercent-io-skill-standardization
- Manifest: https://skillstore.pages.dev/api/skills/supercent-io-skill-standardization/manifest

## Capabilities

- Validates SKILL.md frontmatter fields such as name, description, allowed-tools, and compatibility.
- Checks name format, directory-name matching, description length, and trigger phrasing.
- Reports missing recommended sections including examples, best practices, and references.
- Provides templates for new SKILL.md files and evals/evals.json test cases.
- Runs batch validation for skill directories that contain SKILL.md files.

## Use Cases

- Prepare a Skill for Publishing: Validate structure, required frontmatter, and recommended sections before submitting a skill.
- Audit a Skill Repository: Batch-check many skill directories and collect errors and warnings for cleanup.
- Improve Skill Activation: Rewrite descriptions so agents understand capabilities, trigger conditions, and related user wording.

## Prompt Templates

### Validate One Skill

```
Use skill-standardization to check the skill in <path>. Summarize errors, warnings, and recommended edits.
```

### Improve a Description

```
Review <path>/SKILL.md. Rewrite the frontmatter description to explain capabilities, trigger conditions, and synonyms while staying under 1024 characters.
```

### Convert Legacy Structure

```
Convert the skill at <path> to standard Agent Skills structure. Preserve intent, normalize headings, and add missing recommended sections.
```

### Run a Release Audit

```
Batch validate all skills under <path>. Prioritize blocking errors, description quality, eval coverage, and scripts that need manual review.
```

## Limitations

- The shell validator uses grep, sed, and awk parsing rather than a full YAML parser.
- It cannot prove that a description will always trigger correctly in every agent client.
- It does not perform deep security scanning of scripts or referenced resources.
- It depends on local filesystem access and a POSIX shell.

## Best Practices

- Run validation before publishing or changing a skill description.
- Keep SKILL.md concise and move long reference material to separate resources.
- Add evaluation prompts that match real user requests and expected outcomes.

## Anti Patterns

- Do not use vague descriptions that only name the domain.
- Do not hide required trigger details in references instead of frontmatter.
- Do not rely on interactive scripts because agent shells may be non-interactive.

## Security Audit

- Audited at: 2026-07-07T05:48:05.073\+00:00
- Summary: Context review found that the static findings are false positives from shell arithmetic, fixed local parsing commands, Markdown backticks, and documentation links. No prompt injection, data exfiltration intent, unsafe network behavior, or arbitrary command execution was found in the reviewed files.

## Stats

- Views: 106
- Downloads: 13
- Favorites: 0
- Popularity score: 0