{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_jutka7qm8y2b","handle":"gemini-deepsearch-mcp","url":"https://wellknown.network/agents/gemini-deepsearch-mcp","links":{"self":"https://wellknown.network/agents/gemini-deepsearch-mcp/record.json","html":"https://wellknown.network/agents/gemini-deepsearch-mcp","markdown":"https://wellknown.network/agents/gemini-deepsearch-mcp/record.md","api":"https://wellknown.network/api/v1/agents/gemini-deepsearch-mcp","status":"https://wellknown.network/api/v1/agents/gemini-deepsearch-mcp/status","claim":"https://wellknown.network/agents/gemini-deepsearch-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/gemini-deepsearch-mcp/claim.json","badge":"https://wellknown.network/agents/gemini-deepsearch-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/gemini-deepsearch-mcp/history","tools":"https://wellknown.network/api/v1/agents/gemini-deepsearch-mcp/tools"},"ard":{"identifier":"urn:air::server:gemini-deepsearch-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"gemini-deepsearch-mcp","summary":"Gemini DeepSearch MCP - Automated research agent with Google Gemini models","description":"# Gemini DeepSearch MCP\n\nGemini DeepSearch MCP is an automated research agent that leverages Google Gemini models and Google Search to perform deep, multi-step web research. It generates sophisticated queries, synthesizes information from search results, identifies knowledge gaps, and produces high-quality, citation-rich answers.\n\n## Features\n\n- **Automated multi-step research** using Gemini models and Google Search\n- **FastMCP integration** for both HTTP API and stdio deployment\n- **Configurable effort levels** (low, medium, high) for research depth\n- **Citation-rich responses** with source tracking\n- **LangGraph-powered workflow** with state management\n\n## Usage\n\n### Development Server (HTTP + Studio UI)\nStart the LangGraph development server with Studio UI:\n```bash\nmake dev\n```\n\n### Local MCP Server (stdio)\nStart the MCP server with stdio transport for integration with MCP clients:\n```bash\nmake local\n```\n\n### Testing\nRun the test suite:\n```bash\nmake test\n```\n\nTest the MCP stdio server:\n```bash\nmake test_mcp\n```\n\nUse MCP inspector\n```bash\nmake inspect\n```\n\nWith Langsmith tracing\n```bash\nGEMINI_API_KEY=AI******* LANGSMITH_API_KEY=ls******* LANGSMITH_TRACING=true make inspect\n```\n\n## API\n\nThe `deep_search` tool accepts:\n- **query** (string): The research question or topic to investigate\n- **effort** (string): Research effort level - \"low\", \"medium\", or \"high\"\n  - **Low**: 1 query, 1 loop, Flash model\n  - **Medium**: 3 queries, 2 loops, Flash model  \n  - **High**: 5 queries, 3 loops, Pro model\n\n### Return Format\n\n**HTTP MCP Server** (Development mode):\n- **answer**: Comprehensive research response with citations\n- **sources**: List of source URLs used in research\n\n**Stdio MCP Server** (Claude Desktop integration):\n- **file_path**: Path to a JSON file containing the research results\n\nThe stdio MCP server writes results to a JSON file in the system temp directory to optimize token usage. The JSON file contains the same `answer` and `sources` data as the HTTP version, …","publisher":null,"homepage":"https://github.com/alexcong/gemini-deepsearch-mcp","repository":"https://github.com/alexcong/gemini-deepsearch-mcp","version":"0.1.5","license":"MIT","protocols":["mcp"],"tags":["mcp","gemini","research","ai","langgraph"],"pricing":null,"endpoints":[{"url":"pypi:gemini-deepsearch-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"research.web","name":"Deep Web Research","confidence":0.905,"provenance":"derived"}],"categories":["research"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"gemini-deepsearch-mcp","registry":"pypi","observedAt":"2026-11-03T03:26:49.707Z","publishedAt":"2025-06-22T00:16:37.596962Z","latestVersion":"0.1.5"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"gemini-deepsearch-mcp","url":"https://pypi.org/project/gemini-deepsearch-mcp/","firstSeenAt":"2026-09-09T16:23:28.842Z","fetchedAt":"2026-11-03T03:25:57.375Z","normalizedAt":"2026-11-03T03:25:57.375Z"}]},"firstSeenAt":"2026-09-09T16:23:28.842Z","updatedAt":"2026-11-03T03:26:49.707Z"}