AI & Machine Learning

chiark.ai

Discovers and evaluates AI agents for task routing and performance tracking.

1 endpoint5 known toolsFirst detected May 12, 2026Last detected June 7, 2026

ENDPOINT 1

https://chiark.ai/mcp/%60

No auth detected

Find the best AI agents for a given task.

Inferred read-only

Get the full quality score breakdown for a named AI agent.

Inferred read-only

Check the real-time status of an agent.

Inferred read-only

Report the outcome of routing a task to an agent.

Inferred read-only

Get ecosystem-wide statistics for all indexed AI agents.

Inferred read-only

CONNECT WITH APPROVAL

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.chiark-ai]
url = "https://chiark.ai/mcp/%60"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "chiark-ai": {
      "type": "http",
      "url": "https://chiark.ai/mcp/%60"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: chiark-ai
Remote MCP URL: https://chiark.ai/mcp/%60

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "chiark-ai": {
      "url": "https://chiark.ai/mcp/%60"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "chiark-ai": {
      "type": "http",
      "url": "https://chiark.ai/mcp/%60"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "chiark-ai",
  "transport": "streamable-http",
  "url": "https://chiark.ai/mcp/%60"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

BuiltWith Trust API v2 evidence for chiark.ai was fetched 2026-09-14T01:02:26.030Z.

Trust status Neutral

chiark.ai is assessed as Neutral: No suspicious signals found, but no strong positive signal either

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.