okx__agent-skills/CONTRIBUTING.md

Contributing to agent-skills

Thank you for your interest in contributing! This repository welcomes skill additions, corrections, and improvements.

What is a skill?

A skill is a single Markdown file (SKILL.md) that teaches an AI agent how to use a CLI tool for a specific category of tasks. Each skill lives in its own subdirectory under skills/.

How to add a new skill

1. Create a directory: skills/<skill-name>/
2. Create skills/<skill-name>/SKILL.md with the required frontmatter (see below)
3. Open a pull request with a clear description of what the skill covers

Required frontmatter

---
name: skill-name
description: "One-sentence description used by the agent routing system. List the natural-language phrases and scenarios this skill handles."
license: MIT
metadata:
  author: your-name-or-org
  version: "1.0.0"
  agent:
    requires:
      bins: ["cli-tool-name"]
---

The description field is critical — it is parsed by the agent to decide which skill to activate. Be specific and enumerate realistic user phrases.

Skill document structure

A well-formed skill document includes:

• Prerequisites — install instructions and credential setup
• Demo vs Live Mode — if the CLI supports simulated trading, explain the flag
• Skill Routing — clarify what this skill does and does not cover, and point to related skills
• Quickstart — 3–5 working commands a user can run immediately
• Command Index — table of all commands with READ/WRITE classification
• Operation Flow — step-by-step decision guide for common tasks
• CLI Command Reference — parameter tables for each command
• MCP Tool Reference — map CLI commands to underlying tool names
• Input / Output Examples — realistic user requests with expected commands and outputs
• Edge Cases — common pitfalls and how to handle them

How to improve an existing skill

• Fix incorrect commands or parameter names against the actual CLI source
• Add missing commands that exist in the CLI but are not documented
• Improve description trigger phrases for better agent routing
• Add new cross-skill workflow examples

Pull request checklist

• Skill frontmatter is complete and valid YAML
• All CLI commands have been verified against the actual CLI binary
• WRITE commands are clearly marked and include safety confirmation notes
• No internal URLs, credentials, or personally identifiable information included
• license: MIT is set in frontmatter

For reviewers

Before approving a skill MR, reviewers should follow the Skill Review Checklist.

Reporting issues

Open a GitHub issue for:

• Incorrect or outdated CLI commands
• Missing skills for supported CLI modules
• Routing conflicts between skills

License

By contributing, you agree that your contributions will be licensed under the MIT License.