MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
mcp-research-agentgithub.com/amannagarkar/mcp-research-agentupdated this year
—
-
research-agent-mcpgithub.com/gokuno1/research-agent-mcpupdated this year
—
-
Central async research commons for persistent AI agents: identity, claims, tasks, evidence, reputation, MCP.
-
investment-research-agentgithub.com/sarbaniAi/investment-research-agentupdated this year
Multiagent Investment Research Agent built with Databricks Custom MCP Server & Databricks-Langgraph Tool Calling Agent Framework
-
Multi-agent LLM research platform: parallel data agents, synthesis agents, deterministic guardrails, MCP server
MIT ★ 1 -
Evidence-first research workbench: bounded agents, audited citations, offline fixtures, live adapters, MCP and measured evaluations.
MIT -
foundry-medical-research-agentsgithub.com/nrs2130/foundry-medical-research-agentsupdated this year
Medical Research MCP Agents for Azure AI Foundry - PubMed, ClinicalTrials.gov, OpenFDA, and UniProt servers
-
reddit-research-agentgithub.com/soumyadeep-ux/reddit-research-agentupdated this year
MCP server wrapping Reddit API for researching resume/career subreddits. Human-in-the-loop community engagement.
★ 1 -
Read-only gates of autonomous trading agents with deterministic execution: cost, data, validation.
pypi -
research-navigator-agentgithub.com/maryam-na/research-navigator-agentupdated this year
Local-first research discovery agent with ADK-facing orchestration, MCP-ready tools, evidence grounding, and Streamlit review dashboard.
-
PaperPilot-MCP-agentic-research-server-github.com/luharukas/PaperPilot-MCP-agentic-research-server-updated this year
Demo MCP server + Python clients with paper indexing and prompt/resource utilities
-
research-agent-mcp-by-nova-civaigithub.com/civai-technologies/research-agent-mcp-by-nova-civaiactive
research-agent MCP server by Nova (CIVAI), remote Streamable HTTP
MIT -
Multi-agent AI system for automated deep web research using CrewAI, LinkUp, and Deepseek R1 with Streamlit UI and MCP server integration.
MIT -
marketintel-research-agent-n8ngithub.com/chait6982/marketintel-research-agent-n8nupdated this year
n8n research agent that calls a custom MCP tool server to produce sourced competitor and market analysis from a single chat prompt.
-
Paper research assistant with hybrid RAG, source-grounded QA, field-level evidence and tool-calling agents. 多源检索、可溯源问答与跨论文比较(Web / CLI / MCP)。
-
家用储能选型与报价:公开套餐溯源、LangGraph 研究与恢复、pgvector/HNSW 权限检索、JWT/MCP、SSE 与人工审核。
MIT -
Local-first AI/ML research curation: Quick Search, Learning Path, Research Lineage, Trending Feed, My Feed, Venues. Data from Semantic Scholar, HuggingFace, GitHub. MCP server included.
MIT ★ 2 -
銘柄コードから株価・財務・ニュースを集め、8セクションのHTML株式リサーチレポートを自動生成するエージェント(MCP × Claude Code / LLMなしのCLI版あり) | MCP-based equity research agent — three FastMCP servers, Claude Code skills, and a standalone CLI that produce 8-section interactive HTML research reports.
-
finance-research-agentgithub.com/sreenathvemula/finance-research-agentupdated this year
AI equity-research analyst for Indian equities (NSE/BSE) - Claude Agent SDK + 28 MCP tools + RAG + forensic/valuation/SWOT skills. Not investment advice.
Apache-2.0 -
Live consumer demand data for AI market research agents. Amazon, Google Shopping, TikTok, Reddit, App Store. Free tier.
MIT -
Deep-research-agentgithub.com/jakkapat-kingthong/Deep-research-agentupdated this year
Production-grade agentic research system with schema-enforced verifiable citations to eliminate AI hallucination. Built with LangGraph & FastMCP.
MIT ★ 2 -
Research memory for AI agents — index arXiv/S2/PDF papers into SQLite, find coverage gaps, file survey reports. Rust CLI + 18-tool MCP server, also a 5-host plugin.
Apache-2.0 ★ 2 -
LLM agent and MCP server that audits model code for look-ahead and selection bias, scored on a planted-bug eval suite.
MIT -
Read-only Reddit research MCP server. Finds the subreddits where a topic is actually discussed, ranks threads by real conversation instead of upvotes, and returns verbatim quotes tagged by intent with a permalink on every one.
MIT -
ai-researcher-agentgithub.com/evanyfyip/ai-researcher-agentupdated this year
AI Researcher Agent: a LangChain-based MCP tool that aggregates web, ArXiv, and custom scraper results, summarizes them with an LLM, and produces styled HTML/Markdown reports (with banners, collapsible source cards, and JSON artifacts) for pulse and targeted research.
-
4-agent MCP system with LangGraph, RAG, and tool/function calling. Improves faithfulness +6.4% and completeness +21.5% over single-agent baseline.
-
LangGraph multi-agent backend with planner, coordinator, curator, writer and reviewer loop, typed artifacts, MCP tools and traceable runs.
MIT -
Agentic RAG research assistant: a LangGraph ReAct agent queries your own documents through a custom MCP server (ChromaDB) and falls back to Tavily web search when the knowledge base lacks the answer.
-
Read-only Reddit research MCP server. Finds the subreddits where a topic is actually discussed, ranks threads by real conversation rather than upvotes, harvests comments, and returns verbatim quotes tagged by intent with a permalink on every one. Works wi
npm MIT -
Research digest agent built twice from one toolset — as an MCP server any agent can plug into, and as a standalone tool-using Claude agent — with deduplicated storage and a read-only dashboard
MIT -
ai-research-agent-mcpgithub.com/prabureddy/ai-research-agent-mcpupdated this year
🚀 Autonomous AI Research Engineer powered by MCP | Give it a task, watch it research the web, query your knowledge base, write & execute code, generate visualizations, and produce comprehensive reports with self-evaluation | Built for Claude Desktop | Python 3.10+ | RAG + Web Scraping + Code Sandbox | 5-min setup ⚡
MIT ★ 17 -
agentic-research-engine-ossgithub.com/TheAiSingularity/agentic-research-engine-ossupdated this year
The best $0 research agent that runs on a laptop. Local Gemma 3 4B + SearXNG + trafilatura. CLI · TUI · Web GUI · MCP server · submittable Claude plugin. 264 tests, all mocked.
pypi MIT ★ 2 -
seo-keyword-research-for-agents-mcpgithub.com/CleanData-Labs/seo-keyword-research-for-agents-mcpactive
Keyword research as a remote MCP tool for AI agents (Claude, Cursor, VS Code) and via the Apify API: Google Autocomplete ideas and questions plus Google Trends interest and rising queries. No API key. Python and Node examples.
MIT -
Probe Research: experiment tracking for coding agents. Plugins and skills for Claude Code, Codex and Cursor plus a read-only MCP server, so agents build on what your team already tried. CLI: pip install probe-research / npx probe-research. CI-generated mirror of Probe's monorepo.
go Apache-2.0 -
A LangGraph multi-agent system where 4 AI agents autonomously research any topic, extract insights, write a report, and self-review with a conditional revision loop — powered by Groq (LLaMA 3.3 70B), FastAPI, React, and Supabase.
★ 3 -
Three-agent research system on the OpenAI Agents SDK — Planner/Executor/Reviewer with typed handoffs, a custom MCP server, function tools, input/output guardrails, token-streaming (CLI + SSE), tracing, and an LLM-judge eval harness with a pytest regression gate.
-
Safeguarded AI research agent: Claude Code runs headless with one tool, a custom MCP fetch gate that masks personal information, while code checks every answer and a person reviews the results. Built for a local business in the Vic Park AI Business Boost program.
MIT -
Research methods, workflow tools, and independent review for AI agents, packaged as a Skill, plugin, and local MCP server.
MIT ★ 314 -
Open-source deep research MCP. Qwen3-30B-A3B-Thinking via OpenRouter, cited web synthesis for Claude Code, Codex, Cursor, Hermes and any MCP-compatible agent.
MIT ★ 66 -
Parallel multi-agent deep research engine with confidence-scored citations. CLI, Python library and MCP server (install from source).
Apache-2.0 ★ 1 -
Claude Research-style MCP server implementing Anthropic's multi-agent research methodology
npm MIT ★ 2 -
OpenRouter MCP server for research agents
npm MIT -
@iflow-mcp/openrouter-deep-research-mcpupdated this year
OpenRouter MCP server for research agents
npm MIT -
MCP Deep Research Server using Gemini creating a Research AI Agent
MIT ★ 70 -
Deep research in Rust: 5 source types, multi-agent belief debate, SQLite-backed queue, MCP server for OpenCode
MIT ★ 1 -
@raysker/axon-researchgithub.com/raysker/axon-research-mcpupdated this year
Axon's research capabilities as MCP tools - AI agent ecosystem research, DeFi analysis, and more
npm MIT -
Add ChatGPT-style deep research to your agent or product. Self-hosted multi-agent research API with async jobs, signed webhooks, SSE streaming, MCP server, and 7 LLM providers.
MIT ★ 1 -
Deep research stack for OpenClaw agents: 7 MCP servers, 6 audited skills, 1 orchestrator. One-command install.
MIT -
MCP server implementing Anthropic's multi-agent research architecture — parallel subagents, adaptive stopping, citations. On npm.
npm MIT -
@iflow-mcp/wheattoast11-openrouter-deep-research-mcpupdated this year
Production MCP server for multi-agent AI research with OpenRouter
npm MIT -
🔬 Local SQLite-based AI research agent swarm with GOAP planning, multi-perspective analysis, long-horizon recursive framework, AgentDB self-learning, anti-hallucination controls, and MCP server. Goal-oriented with parallel execution. No cloud dependencie
npm ISC -
Research infrastructure for AI agents: papers, protocol specifications and cited source passages through MCP.
AGPL-3.0 ★ 1 -
arxiv-research-mcpgithub.com/jananiv07/arxiv-mcp-serverupdated this year
An MCP server exposing arXiv research tools (search, abstracts, author lookup, trending) to LLM agents.
pypi MIT -
ai-research-assistantgithub.com/aabhas-gupta/ai-research-assistantupdated this year
Full-stack AI Research Assistant — RAG, Agents, MCP, ChromaDB, Streamlit
NOASSERTION -
Production-oriented multi-agent research intelligence platform combining LangGraph orchestration, Model Context Protocol (MCP), hybrid vector retrieval, claim verification, and evidence synthesis.
-
gemini-researchergithub.com/capybearista/gemini-researcher-mcpupdated this year
Stateless MCP server that proxies research queries to Gemini CLI, reducing agent context/model usage
npm BSD-3-Clause -
Local, auditable market research for AI agents with exact evidence provenance, claim verification, counterevidence, MCP, and deterministic evidence packs.
MIT ★ 1 -
research-mcpgithub.com/OrbisAPI/research-mcpupdated this year
500+ research, web search, scraping & data extraction APIs for AI agents. Pay per call with USDC via x402.
★ 5 -
MCP server for **article/tutorial research during agent planning** — dev.to search, deduped ranked briefs, methodology aligned with Hermes `deep-research`.
MIT ★ 1 -
@iflow-mcp/reka-ai-reka-research-mcpupdated this year
Reka Research MCP Server - Agentic analysis tools including fact-checking, claim verification, and similarity search
npm ISC