Docs

How to get the data: hand it to your agent, call the API, or find the dataset you need in the catalog.

AI agents and MCP
Paste one line into Claude Code, Codex or Cursor and the agent sets itself up, or connect Claude, ChatGPT or Cursor to the MCP server.
- MCP server
- llms.txt
Open /agents →
paste into your agent
```
Read https://aperiodic.io/start.md and follow it.
```
Open inClaude Desktop Claude Code (terminal)Codex
Or add it toClaude ChatGPT Cursor VS Code
Open /agents →
API reference
History as Parquet over REST and every new row over WebSocket, on one key, with the Python SDK and CLI that wrap them.
Open /api-docs →
Data catalog
Every dataset and field, with a sample of each you can pull without a key.
Open /catalog →

How to get the data: hand it to your agent, call the API, or find the dataset you need in the catalog.

AI agents and MCP
Paste one line into Claude Code, Codex or Cursor and the agent sets itself up, or connect Claude, ChatGPT or Cursor to the MCP server.
- MCP server
- llms.txt
Open /agents →
paste into your agent
```
Read https://aperiodic.io/start.md and follow it.
```
Open inClaude Desktop Claude Code (terminal)Codex
Or add it toClaude ChatGPT Cursor VS Code
Open /agents →
API reference
History as Parquet over REST and every new row over WebSocket, on one key, with the Python SDK and CLI that wrap them.
Open /api-docs →
Data catalog
Every dataset and field, with a sample of each you can pull without a key.
Open /catalog →

AI agents and MCP