MCP server that provides Google Search via Gemini's built-in Grounding with Google Search feature.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-gemini-google-search","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# mcp-gemini-google-search An MCP stdio server for Google-grounded Gemini answers and asynchronous Gemini Deep Research reports. This repository ports the behavior of [`yukukotani/mcp-gemini-google-search`](https://github.com/yukukotani/mcp-gemini-google-search) and extends it with Deep Research. The upstream project publishes the `mcp-gemini-google-search` npm package; this Python port shares the name but installs from this Git repository rather than from a package index. The Python implementation uses the official [`google-genai`](https://googleapis.github.io/python-genai/) SDK for Gemini API and Vertex AI access and the official [MCP Python SDK](https://github.com/modelcontextprotocol/python-sdk) low-level server. ## Features - Always advertises three standard tools: `google_search`, `deep_research`, and `deep_research_result` - Runs Google Search grounding through the [Interactions API](https://ai.google.dev/gemini-api/docs/interactions) - Optionally enables [URL context](https://ai.google.dev/gemini-api/docs/interactions/url-context) and [code execution](https://ai.google.dev/gemini-api/docs/interactions/code-execution) per server or per search request - Sends every `google_search` as an independent interaction with `store=false` - Starts Deep Research as a provider-managed background interaction and returns a durable `interaction_id` immediately - Returns readable text content plus schema-backed `structuredContent`; failures are readable MCP tool errors rather than opaque protocol errors - Formats successful answers and reports as GitHub-flavored Markdown with inline `[n]` markers and a linked `## Sources` list - Supports Google AI Studio and Vertex AI ## Choose a tool | Need | Tool | Behavior | | ------------------------------------------------------------------ | --------------------…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.