agentic-research

MCP ServerMCP Registry官方收录

io.github.TheAiSingularity/agentic-research · v0.1.3

Local research agent that verifies its own answers. Runs on Gemma 3 4B + Ollama, $0/query.

源仓库 ↗

概览

agentic-research 是一个MCP Server，收录自官方 MCP Registry。支持 stdio 传输。本页提供 Cursor、Claude Code 等客户端的安装配置片段。

安装

选择你的平台查看安装方式

{
  "mcpServers": {
    "agentic-research": {
      "command": "agentic-research-mcp",
      "args": [
        "agentic-research-engine"
      ]
    }
  }
}

环境变量

OPENAI_BASE_URL可选

Any OpenAI-compatible endpoint. Default: OpenAI cloud. Use http://localhost:11434/v1 for Ollama.

OPENAI_API_KEY可选secret

API key for the endpoint above. Use 'ollama' as a sentinel value when running locally against Ollama.

MODEL_SYNTHESIZER可选

Model identifier used for the synthesize node. Defaults to 'gpt-5-mini'; set to 'gemma3:4b' for Mac-local Ollama.

MODEL_PLANNER可选

Model for the plan / classify / critic / compress / verify nodes. Defaults to 'gpt-5-nano'.

EMBED_MODEL可选

Embedding model identifier (for retrieval + memory). Default 'text-embedding-3-small'; use 'nomic-embed-text' on Ollama.

SEARXNG_URL可选

Base URL of the SearXNG meta-search instance. Default http://localhost:8888.

LOCAL_CORPUS_PATH可选

Path to an index directory built via scripts/index_corpus.py. When set, local corpus hits augment web search.

ENABLE_RERANK可选

Set to '1' to enable the BAAI/bge-reranker-v2-m3 cross-encoder rerank stage. First run downloads ~560MB.

ENABLE_FETCH可选

Set to '0' to skip the trafilatura full-page fetch stage. Default '1'.

统一 Manifest

{
  "id": "io.github.TheAiSingularity/agentic-research",
  "type": "mcp-server",
  "version": "0.1.3",
  "displayName": "agentic-research",
  "description": "Local research agent that verifies its own answers. Runs on Gemma 3 4B + Ollama, $0/query.",
  "repository": {
    "url": "https://github.com/TheAiSingularity/agentic-research-engine-oss",
    "source": "github"
  },
  "distribution": {
    "packages": [
      {
        "registryType": "pypi",
        "identifier": "agentic-research-engine",
        "version": "0.1.3",
        "runtimeHint": "agentic-research-mcp",
        "transport": "stdio",
        "environmentVariables": [
          {
            "name": "OPENAI_BASE_URL",
            "description": "Any OpenAI-compatible endpoint. Default: OpenAI cloud. Use http://localhost:11434/v1 for Ollama."
          },
          {
            "name": "OPENAI_API_KEY",
            "description": "API key for the endpoint above. Use 'ollama' as a sentinel value when running locally against Ollama.",
            "isSecret": true
          },
          {
            "name": "MODEL_SYNTHESIZER",
            "description": "Model identifier used for the synthesize node. Defaults to 'gpt-5-mini'; set to 'gemma3:4b' for Mac-local Ollama."
          },
          {
            "name": "MODEL_PLANNER",
            "description": "Model for the plan / classify / critic / compress / verify nodes. Defaults to 'gpt-5-nano'."
          },
          {
            "name": "EMBED_MODEL",
            "description": "Embedding model identifier (for retrieval + memory). Default 'text-embedding-3-small'; use 'nomic-embed-text' on Ollama."
          },
          {
            "name": "SEARXNG_URL",
            "description": "Base URL of the SearXNG meta-search instance. Default http://localhost:8888."
          },
          {
            "name": "LOCAL_CORPUS_PATH",
            "description": "Path to an index directory built via scripts/index_corpus.py. When set, local corpus hits augment web search."
          },
          {
            "name": "ENABLE_RERANK",
            "description": "Set to '1' to enable the BAAI/bge-reranker-v2-m3 cross-encoder rerank stage. First run downloads ~560MB."
          },
          {
            "name": "ENABLE_FETCH",
            "description": "Set to '0' to skip the trafilatura full-page fetch stage. Default '1'."
          }
        ]
      }
    ],
    "remotes": []
  },
  "dependencies": [],
  "installTargets": [
    "claude-code",
    "claude-desktop",
    "cursor",
    "vscode"
  ],
  "keywords": [],
  "provenance": {
    "origin": "official-mcp-registry",
    "originalId": "io.github.TheAiSingularity/agentic-research",
    "originalUrl": "https://registry.modelcontextprotocol.io/v0.1/servers/io.github.TheAiSingularity%2Fagentic-research/versions/latest",
    "isOfficial": true,
    "status": "active"
  }
}

agentic-research

概览

安装

环境变量

相关资源

RFP.ai

xenarch-mcp

api

agent-lsp

hostinger-api-mcp

Unterm

统一 Manifest