§ MCP · Model Context Protocol

Give an agent the whole corpus

Every number and quote in this app comes from a live query — and the same data is exposed to AI agents over a Model Context Protocol server. Add it as a connector and your agent can search, navigate the topic web, profile any watchdog and pull verbatim evidence. It’s public and read-only.

Add it as a connector

Connector URLhttps://aicolab-audit-corpus.aicolab.workers.dev/mcp

In Claude, open Settings → Connectors (or a project’s connectors) and choose Add custom connector.
Paste the URL above and save. No authentication is required — the tools are public and read-only.
Start a chat and ask something like “Using the audit corpus, which oversight body grades hardest on cyber, and quote them.” The agent will call the tools below.

Works with any MCP client (Claude Desktop, Claude Code, the API). The endpoint speaks JSON-RPC over a stateless one-POST Streamable-HTTP adapter.

Try the tools

Pick a tool, adjust the inputs, and run it against the live server — the exact request an agent would make. Results are the raw tool output.

Overview & slicing

Find & relate topics

Anatomy of a topic

Reports & bodies

find_topics

Query *Limit

POST /mcp

{
  "method": "tools/call",
  "params": {
    "name": "find_topics",
    "arguments": {
      "query": "solitary confinement of children",
      "limit": "8"
    }
  }
}

Response

Run a tool to see its output.