# Coolhand — Human Feedback for Better AI Outputs

Coolhand is an LLM observability and human feedback platform. It helps engineering teams capture user feedback on AI outputs, log LLM requests across providers, and automatically generate pull requests to improve prompt quality over time.

## What Coolhand Does

- **LLM Request Logging**: Capture every request and response from OpenAI, Anthropic, Google, and other providers through a single ingest endpoint. No SDK changes required for structured logging.
- **Human Feedback Collection**: Embed lightweight feedback widgets (thumbs up/down, corrections, ratings) directly in your product. Feedback is linked to the exact LLM request that generated the output.
- **Automated Prompt Improvement**: Coolhand analyzes feedback patterns and opens pull requests with suggested prompt improvements — closing the loop between user signals and model behavior.
- **Prompt Templates**: Manage and version your prompts as named templates. Track which version of a prompt generated which output.
- **Inference API Catalog**: A live, curated table of pricing and context window specs for 100+ LLM models. Updated as providers change their pricing.

## SDKs

Coolhand provides SDKs for the most common backend languages:

- **Ruby** — `gem "coolhand"`
- **Python** — `pip install coolhand`
- **Node.js** — `npm install coolhand`

Each SDK wraps your existing LLM calls with a single line of code and handles logging, feedback collection, and prompt template management.

## Getting Started

1. Sign up at [coolhandlabs.com](https://coolhandlabs.com)
2. Install the SDK for your language
3. Wrap your LLM calls with the Coolhand logger
4. Embed a feedback widget in your UI
5. Review feedback and apply suggested prompt improvements

## API

Coolhand exposes a REST API for all features. The full API reference is available at [coolhandlabs.com/docs](https://coolhandlabs.com/docs). The machine-readable OpenAPI spec is at [coolhandlabs.com/api-docs/v2/coolhand_api.yaml](https://coolhandlabs.com/api-docs/v2/coolhand_api.yaml).

Authentication uses an API key passed via the `X-API-Key` header.

## MCP Integration

Coolhand exposes a Model Context Protocol (MCP) server at `https://coolhandlabs.com/mcp`. This allows AI coding assistants (Claude Code, Cursor, etc.) to interact directly with your Coolhand workspace — querying logs, managing templates, and reviewing feedback without leaving the assistant.

The public MCP endpoint at `https://coolhandlabs.com/mcp/public` provides access to the inference API catalog without authentication.