Skill

validate-markdown-refs

Validate markdown file references in .claude and .prism directories. Use to find broken links before committing documentation changes.

npx claudepluginhub resolve-io/.prism

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Scan markdown files for broken internal links. Returns clean JSON output.

Supporting Assets

reference/output-format.mdscripts/validate-refs.py

SKILL.md

Similar Skills

agent-payment-x402

170.3k

Enables AI agents to execute x402 payments with per-task budgets, spending controls, and non-custodial wallets via MCP tools. Use when agents pay for APIs, services, or other agents.

everything-claude-code

autonomous-loops

170.3k

Provides patterns for autonomous Claude Code loops: sequential pipelines, agentic REPLs, PR cycles, de-sloppify cleanups, and RFC-driven multi-agent DAGs. For continuous dev workflows without intervention.

everything-claude-code

nestjs-patterns

170.3k

Applies NestJS patterns for modules, controllers, providers, DTO validation, guards, interceptors, config, and production TypeScript backends with project structure and bootstrap examples.

everything-claude-code

Stats

Parent Repo Stars1

Parent Repo Forks2

Last CommitMar 2, 2026

Actions

View Source View Plugin View on GitHub View README

Validate Markdown References

Scan markdown files for broken internal links. Returns clean JSON output.

When to Use

Before committing documentation changes
When the link-checker agent needs to verify its findings
Manual validation of plugin documentation integrity
CI/CD pipelines for documentation quality gates

Quick Start

# Run from project root
python .prism/plugins/prism-devtools/skills/validate-markdown-refs/scripts/validate-refs.py

# Or with custom directories
python validate-refs.py --directories .claude .prism

# Include archive directories
python validate-refs.py --include-archive

What It Checks

Markdown links: [text](path) and [ref]: path syntax
Relative paths: Resolved from source file location

What It Skips

URLs (http://, https://, mailto:)
Anchor-only links (#section)
Links inside fenced code blocks
Template variables ({{...}}, ${...})
plugins/cache directory (duplicates)

Output

Returns JSON to stdout. See output format for schema.

Example:

{
  "status": "FAIL",
  "summary": {
    "files_scanned": 127,
    "broken_links": 3
  },
  "broken_links": [
    {
      "source_file": ".prism/plugins/prism-devtools/skills/README.md",
      "line": 58,
      "target_path": "./orca-local-setup/SKILL.md",
      "error": "File not found"
    }
  ]
}

Exit Codes

Code	Meaning
0	No broken links found
1	Broken links found
2	Script error

Integration with Link-Checker Agent

The link-checker agent can invoke this script to verify its findings:

python validate-refs.py

JSON is the default (and only) output format, providing deterministic validation to complement the agent's AI-based scanning.