mcpindex

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

The MCP server index that vets servers, not just lists them. Advisory screen before you install.

Measured statistics

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "mcpindex": {
      "url": "https://mcpindex.ai/api/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (6)

• recommend_mcp_for_task

  Recommend the best MCP servers for a natural-language task. Returns top 3 ranked picks with reasoning, install commands, and quality scores. Use this when the user asks for the right MCP server for a task they want to do.

  Parameters:

  • task (required) — string — Natural-language description of the task, e.g. "read PDFs and write to S3".

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• search_mcp_servers

  Keyword + semantic search across every MCP server mcpindex indexes. Use when the user knows what tool category they want but not which server.

  Parameters:

  • query (required) — string — Search query.
  • limit — integer — Max results (default 10, max 50).
  • category — string — Optional category filter (e.g. database, browser, github).

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_install_command

  Get the exact install command for a given MCP server and client. Returns a JSON block ready to paste into the client config.

  Parameters:

  • server_slug (required) — string — Slug of the server (from search or recommend results).
  • client (required) — string — Target client.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• compare_servers

  Side-by-side comparison of 2-5 MCP servers - quality scores, install paths, env vars.

  Parameters:

  • slugs (required) — array — Server slugs to compare.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• check_tool_trust

  Pre-invocation advisory screen for a specific tool on an MCP server. Returns an advisory verdict object (directive ALLOW | DENY | REVIEW | UNVERIFIED, dimensions, freshness). At v1 the public screen produces REVIEW or UNVERIFIED only - ALLOW/DENY are reserved. Not the in-path gate (mcpindex-gate). Agents SHOULD treat UNVERIFIED as "human review required", never as ALLOW.

  Parameters:

  • server_id (required) — string — Registry slug from search_mcp_servers / recommend results, e.g. "io-github-microsoft-playwright-mcp" (NOT a short name like "github").
  • tool_name (required) — string — Tool name as exposed by the server (e.g. "create_pull_request").

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• assess_server

  Aggregated pre-flight trust assessment across all tools on an MCP server. Same verdict shape as check_tool_trust. Use for "is THIS server worth integrating?" decisions. v1 advisory; conformance monitored not enforced; verdicts may be UNVERIFIED if not yet probed.

  Parameters:

  • server_id (required) — string — Server slug to assess.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

Publisher website Repository

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.