Coolhand Help Center
How to capture your LLM requests and the human feedback on them, what Coolhand records for each provider, and how that turns into prompt and code fixes you can review.
Provider guides
What Coolhand records for each provider, which fields it can and cannot capture, and how to send requests so your prompts group cleanly into templates.
- TypeSafe Best Practices
- Anthropic API best practices
- OpenAI API best practices
- Azure OpenAI best practices
- AWS Bedrock best practices
Feedback & prompts
How to capture high-signal human feedback on AI output, how Coolhand scores and attributes it, and the prompt patterns that hold across every provider.
- How should I collect feedback on AI output?
- What is feedback match rate?
- How is feedback quality scored?
- What is partial feedback?
- Creators vs. reviewers
Coding agents
How Coolhand ingests Claude Code, Claude Cowork, and GitHub Copilot sessions, and what each transcript format can and cannot tell you.
Logs & API
What every field on an LLM request log means, when batching is worth it, and how to query your observability data from an AI agent over MCP.
- How can I get help articles in Markdown format?
- How do I create or update a template using an external key?
- How do I list and analyze my workloads using the API?
- LLMs and System One models: what's different, and when to use which
- Node.js/JavaScript SDK Integration Best Practices
Integrations
How to connect Coolhand to your source control and the other systems your team already runs, including the permissions, tokens and webhooks each one needs.