Aidemos

Official MCP Registry

A public, read-only Model Context Protocol server that gives any AI agent structured, evidence-backed access to the FutureSmart AI Demos catalogue: AI-tool reviews, rankings ("best X for Y"), how-to use cases, head-to-head comparisons, and toolkits.

Every ranking is backed by a queryable evidence graph — real observation cells (tool × test-scenario × criterion → verdict, score, researcher note, and the actual input/output screenshots that prove it). Agents don't just get opinions; they get the ground truth behind them.

• Endpoint: https://mcp.aidemos.com/mcp
• Transport: Streamable HTTP (JSON-RPC 2.0)
• Auth: none — all data is already-public catalogue content
• Registry: com.aidemos/catalogue in the official MCP Registry

Add it to your client

Claude Code (CLI)

claude mcp add --transport http aidemos https://mcp.aidemos.com/mcp

Cursor

Add to ~/.cursor/mcp.json (or .cursor/mcp.json in a project):

{
  "mcpServers": {
    "aidemos": {
      "url": "https://mcp.aidemos.com/mcp"
    }
  }
}

VS Code / Codex & other clients (via mcp-remote)

For clients that speak stdio, bridge to the remote server:

{
  "mcpServers": {
    "aidemos": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.aidemos.com/mcp"]
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector → URL https://mcp.aidemos.com/mcp.

Tools (16 verbs)

Tool	What it does
search	Keyword / semantic / hybrid search across the whole catalogue
list_use_cases	List how-to use-case guides
list_rankings	List "best X" ranking pages
list_tools	List AI-tool pages
list_compares	List head-to-head comparison pages
list_toolkits	List curated toolkit bundles
list_personas	List personas with per-type page counts
list_categories	List the category vocabulary in use
tools_in_ranking	Ranked tools (rank + badge) for a ranking
rankings_for_tool	Every ranking a given tool appears in
get_persona	Everything tagged to a persona
get_tool	Full tool detail (scores, pricing, review)
get_ranking	Full ranking detail (criteria, verdicts, winner)
get_use_case	Full use-case guide
get_evidence	Query the evidence graph — cells + real artifacts
compare_tools	Evidence-aligned, same-input comparison of two tools

Quick test

curl -s -X POST https://mcp.aidemos.com/mcp \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

About

Built by FutureSmart AI. The catalogue and this MCP server are part of AI Demos — structured, evidence-first intelligence about AI tools.