# Cero Labs
> Agent-native expert escalation API, Python SDK, TypeScript SDK, and MCP server for AI workflows.

## Start Here
- [For Agents](https://cerolabs.ai/for-agents): primary install guide for MCP clients, SDKs, and skill templates.
- [API Docs](https://cerolabs.ai/docs): human-friendly API reference with curl and SDK examples.
- [Machine-readable API Docs](https://api-production-cd65.up.railway.app/docs.md): concise Markdown reference for agents and automated tooling.
- [Python SDK source](https://github.com/os1123/cero-labs/tree/main/sdk/python): install with `pip install "git+https://github.com/os1123/cero-labs.git#subdirectory=sdk/python"`
- [TypeScript SDK on npm](https://www.npmjs.com/package/cerolabs-sdk): `npm install cerolabs-sdk`
- [MCP Server on npm](https://www.npmjs.com/package/cerolabs-mcp): stdio MCP server for Claude Desktop, Cursor, Codex, Cline, and other MCP clients.

## Core Capabilities
- Escalate an uncertain AI task to a verified human expert.
- Receive structured JSON answers validated against a caller-supplied schema.
- Wait synchronously for resolution or receive webhooks asynchronously.
- Use domain-specific expert pools across healthcare, legal, sales, content moderation, operations, and more.
- Meter usage with a pay-per-escalation credit model.

## Important URLs
- Landing: https://cerolabs.ai/
- For agents: https://cerolabs.ai/for-agents
- Pricing: https://cerolabs.ai/pricing
- Templates: https://cerolabs.ai/templates
- Expert portal: https://cerolabs.ai/expert
- API base URL: https://api-production-cd65.up.railway.app
- Domains endpoint: https://api-production-cd65.up.railway.app/domains

## More
- [Extended agent guide](https://cerolabs.ai/llms-full.txt)