Skip to content
ToolCargo

The MCP directory / Open to explore

MCP server directory.
Find your next tool.

Search 3,974 indexed servers from the official MCP Registry and selected vendor documentation. Find remote endpoints and local packages, inspect the publisher’s setup details, then connect in your AI client.

Where will your next task take you?

Free discovery · no account · no API key

Let your agent do the finding.

Paste into your AI agent
Read https://toolcargo.com/agents.txt, then find the MCP servers I need and help me connect them. Ask me what I want to do first.

Or add https://toolcargo.com/mcp as a Streamable HTTP MCP server. Discovery is free and needs no account.

Compare task fit, publisher, permissions and a small acceptance test with the MCP server selection guide.

Connect directly to official MCP servers

Vendor setup guides for your existing accounts. These connections run with the vendor; ToolCargo does not host them or supply their credentials.

Browse the remote MCP server directory →

Not sure which connection to choose? Read local vs remote MCP servers. Building a discovery workflow? See the MCP server search API reference.

google-surf-mcp

Close details

Web, academic and code search with graph RAG, data lineage, ontology and cross-project schema links.

Directory identity: io.github.HarimxChoi/google-surf-mcp · Version: 1.1.3

Publisher connection metadata
{
  "remotes": [],
  "packages": [
    {
      "registryType": "npm",
      "identifier": "google-surf-mcp",
      "version": "1.1.3",
      "transport": {
        "type": "stdio"
      },
      "environmentVariables": [
        {
          "description": "SearchApi API key. Required only for SearchApi primary or fallback modes.",
          "format": "string",
          "isSecret": true,
          "name": "SEARCH_API"
        },
        {
          "description": "Provider for search and search_parallel: browser, searchapi, or fallback. Defaults to browser.",
          "format": "string",
          "name": "SURF_SEARCH_PROVIDER"
        },
        {
          "description": "Provider for scholar_search: browser, searchapi, or fallback. Defaults to browser.",
          "format": "string",
          "name": "SURF_SCHOLAR_PROVIDER"
        },
        {
          "description": "Browser engine: auto, native, or playwright. Auto uses one hidden native Chrome broker shared across local MCP sessions.",
          "format": "string",
          "name": "SURF_BROWSER_ENGINE"
        },
        {
          "description": "Absolute path to the Chrome binary. Auto-detected on Windows/macOS/Linux when omitted.",
          "format": "string",
          "name": "CHROME_PATH"
        },
        {
          "description": "Directory for the warm Chrome profile. Defaults to ~/.google-surf-mcp.",
          "format": "string",
          "name": "SURF_PROFILE_ROOT"
        },
        {
          "description": "Enable local project memory and hybrid retrieval. Defaults to true.",
          "format": "string",
          "name": "SURF_RESEARCH"
        },
        {
          "description": "Shared route for search and search_parallel: live or hybrid. Defaults to hybrid.",
          "format": "string",
          "name": "SURF_RETRIEVAL_MODE"
        },
        {
          "description": "Embedded SurrealDB data directory. Defaults to <profile>/research.",
          "format": "string",
          "name": "SURF_RESEARCH_ROOT"
        },
        {
          "description": "Local model for HNSW retrieval and shared reranking. Defaults to Xenova/multilingual-e5-small; use off to disable.",
          "format": "string",
          "name": "SURF_RESEARCH_VECTOR_MODEL"
        },
        {
          "description": "Disable ONNX CPU memory arena and memory pattern. Defaults to true.",
          "format": "string",
          "name": "SURF_RESEARCH_VECTOR_LOW_MEMORY"
        },
        {
          "description": "ONNX intra-op thread count from 1 to 16. Defaults to 4.",
          "format": "string",
          "name": "SURF_RESEARCH_VECTOR_THREADS"
        },
        {
          "description": "Sparse-index at most one eligible small GitHub repository per search call. Defaults to true.",
          "format": "string",
          "name": "SURF_RESEARCH_REPO_AUTO"
        },
        {
          "description": "Maximum searchable repository source-text size in MB. Assets are excluded. Defaults to 20.",
          "format": "string",
          "name": "SURF_RESEARCH_REPO_AUTO_MAX_MB"
        },
        {
          "description": "Maximum repository source file count for automatic indexing. Defaults to 2000.",
          "format": "string",
          "name": "SURF_RESEARCH_REPO_AUTO_MAX_FILES"
        },
        {
          "description": "Shared research broker idle lifetime in milliseconds. Defaults to 60000.",
          "format": "string",
          "name": "SURF_RESEARCH_BROKER_IDLE_MS"
        },
        {
          "description": "Maximum concurrent research broker reads. Defaults to 4.",
          "format": "string",
          "name": "SURF_RESEARCH_READ_CONCURRENCY"
        },
        {
          "description": "Timeout per embedded SurrealDB query in milliseconds. Defaults to 120000; timed-out retrieval lanes are reported as partial.",
          "format": "string",
          "name": "SURF_RESEARCH_QUERY_TIMEOUT_MS"
        },
        {
          "description": "Optional token that raises GitHub API limits for repository inspection.",
          "format": "string",
          "isSecret": true,
          "name": "GITHUB_TOKEN"
        },
        {
          "description": "Browser locale, e.g. en-US.",
          "format": "string",
          "name": "SURF_LOCALE"
        },
        {
          "description": "IANA timezone, e.g. America/New_York. Defaults to system timezone.",
          "format": "string",
          "name": "SURF_TZ"
        },
        {
          "description": "Controls Playwright extraction, compatibility, and recovery paths. Native search hides normal system Chrome and shows it only for CAPTCHA recovery.",
          "format": "string",
          "name": "SURF_HEADLESS"
        },
        {
          "description": "Idle ms before closing the sequential ctx and pool. 0 disables idle auto-close. Defaults to 30000.",
          "format": "string",
          "name": "SURF_IDLE_CLOSE_MS"
        },
        {
          "description": "Set to 'true' to allow extract on private/loopback addresses (localhost, 10.x, 192.168.x, 169.254.x, etc). Default blocks them as an SSRF guard.",
          "format": "string",
          "name": "SURF_ALLOW_PRIVATE"
        },
        {
          "description": "Enable the optional workflow-only LLM selector repair path. Defaults to false.",
          "format": "string",
          "name": "SURF_LLM_HEAL"
        },
        {
          "description": "LLM selector repair provider: anthropic or orcarouter. Defaults to anthropic.",
          "format": "string",
          "name": "SURF_LLM_PROVIDER"
        },
        {
          "description": "Optional model override. OrcaRouter defaults to orcarouter/auto.",
          "format": "string",
          "name": "SURF_LLM_MODEL"
        },
        {
          "description": "Anthropic API key for optional LLM-assisted selector repair.",
          "format": "string",
          "isSecret": true,
          "name": "ANTHROPIC_API_KEY"
        },
        {
          "description": "OrcaRouter API key for optional LLM-assisted selector repair.",
          "format": "string",
          "isSecret": true,
          "name": "ORCAROUTER_API_KEY"
        }
      ]
    }
  ]
}

Remote URLs go into your client's MCP settings. Local packages need a compatible runtime. Check required headers, environment variables, permissions and provider costs before setup.

Publisher repository ↗

3,974 results

Index updated 2026-10-02

This directory combines dated registry listings and selected vendor-documentation entries. Documentation checks are not live account or client verification. Third-party servers run with their own providers, credentials and pricing. ToolCargo discovery supplies setup metadata; it does not run their tools. Explore ToolCargo's hosted connectors.