SkillSpec Converter

SkillSpec Converter

Define a skill once and export runtime-ready formats for OpenClaw, Claude, Codex, and MCP manifests.

Validate before exporting:Agent Skill Validator|MCP Server Starter Generator

What This Tool Does

SkillSpec Converter is built for deterministic developer and agent workflows.

Convert one canonical skill definition into OpenClaw SKILL.md, Claude blocks, Codex scaffolds, and MCP manifest snippets.

Use How to Use for execution steps and FAQ for constraints, policies, and edge cases.

Last updated:

This tool is provided as-is for convenience. Output should be verified before use in any production or critical context.

Agent Invocation

Best Path For Builders

Browser workflow

Runs instantly in the browser with private local processing and copy/export-ready output.

Browser Workflow

This tool is optimized for instant in-browser execution with local data handling. Run it here and copy/export the output directly.

/skill-spec-converter/

For automation planning, fetch the canonical contract at /api/tool/skill-spec-converter.json.

How to Use SkillSpec Converter

  1. 1

    Paste canonical skill JSON or YAML

    Provide one canonical skill definition with name, description, trigger, instructions, and optional tools with input schemas.

  2. 2

    Validate structure before conversion

    If required fields are missing, fix the error panel first. Conversion output quality depends on a complete canonical spec.

  3. 3

    Switch target output tabs

    Use OpenClaw, Claude, Codex, and MCP tabs to generate each runtime-specific format from the same source spec.

  4. 4

    Copy each target format

    Use the copy action per target to export SKILL.md blocks, instruction templates, task scaffolds, or MCP manifest snippets.

  5. 5

    Validate converted output

    Run converted specs through your runtime checks or Agent Skill Validator to confirm compatibility before deployment.

Frequently Asked Questions

What is SkillSpec Converter?
SkillSpec Converter transforms one canonical skill definition into multiple runtime-ready outputs: OpenClaw SKILL.md, Claude instruction block, Codex task scaffold, and MCP manifest snippet.
What canonical schema fields are required?
At minimum: name, description, trigger, and instructions. Tools are optional for most outputs but strongly recommended for MCP-ready exports.
Can I use YAML as input?
Yes. The converter accepts both JSON and YAML canonical definitions and generates all target formats from the same source text.
Does this tool store or transmit my skill specs?
No. Conversion runs fully client-side in your browser. Your canonical skill text and generated outputs are not sent to any backend.
How should I verify converted output?
Use Agent Skill Validator for linting and run your normal runtime checks per platform (OpenClaw, Claude, Codex, or MCP) before production rollout.