Skip to main content
The Birdeye Peak MCP server and agent recipes are currently in development and not yet available. This page describes what is planned and how to register interest for early access.

​
About developer AI integration

Peak’s data model is well-suited to AI workflows. The Object + Command structure gives agents a clean vocabulary — every question about a token, protocol, or wallet maps to a specific Command, and every Command returns structured data. This makes Peak data straightforward to integrate into agent reasoning, research assistants, and analytical copilots. Two integration paths are planned.

​
MCP server

The Birdeye Peak MCP server will expose Peak Commands as tools through the Model Context Protocol — the emerging standard for connecting AI agents to external data sources. When available, the MCP server will let AI tools query Peak data the same way the Peak terminal does. Tools like Claude, Cursor, Windsurf, and other MCP-compatible clients will be able to:
  • Look up tokens, protocols, wallets, and other Objects.
  • Run Commands to retrieve structured data.
  • Build on top of Peak’s live datasets without scraping or screen-reading. Specific authentication, tool schemas, and rate limits will be documented when the MCP server becomes available.

​
Agent recipes

Alongside the MCP server, we plan to publish reference patterns for common analytical agents built on top of Peak data. These will cover:
  • Research agents — agents that produce structured token or protocol research using Peak data.
  • Monitoring agents — agents that watch for specific on-chain conditions and surface alerts.
  • Workflow agents — agents that combine Peak data with external sources to handle multi-step analytical tasks. The recipes will include working examples, prompts, and integration code where applicable.

​
Register interest

If you are building agents or AI workflows that would benefit from programmatic Peak access, we would like to hear about your use case. Early access will be offered to teams with concrete integration plans.

Register interest

Email peak@birdeye.so with a short description of what you are building. We will keep you informed as these features approach launch.

​
Looking for AI features today?

While the developer integration paths are in development, two AI surfaces are available right now:

Peak AI

The conversational analyst built into the Peak terminal. Available to all beta users.

Documentation MCP server

The Mintlify-hosted MCP server for this documentation site. Lets AI tools query the docs themselves.

Data API

The programmatic data layer the MCP server will be built on.

Objects and Commands

The data model that will structure the MCP tool surface.