SECRET4422/mcp-deep-research-server

Non-generic MCP for real research. Search → Scrape → Synthesize → Fact-check → Remember.
### Why not generic?
| Generic MCP (boring) | This MCP (pro) |
| echo, fetch | Orchestrated deep research |
| Returns raw HTML | Cheerio + Turndown → clean markdown + headings, links, meta |
| No memory | Persistent memory in ~/.mcp-deep-research/ |
| One page at a time | Parallel 3-worker scraper, 10min cache |
| No reasoning | Fact-check with stance scoring, contradiction detection |
### Architecture
mermaid
graph LR
A[User: deepresearch topic] --> B[searchweb DDG HTML]
B --> C[Parallel Scrape x3-8]
C --> D[cheerio clean + turndown md]
D --> E[extractinsights heuristic]
E --> F[Synthesize Report + Citations]
F --> G[memorysave + history]
F --> H[Return to Claude]
I[comparesources] --> C
J[factcheckclaim] --> B
K[memorysearch] --> G
### Tools (8)
| Tool | What it does | Params |
| searchweb | DuckDuckGo HTML search, no API key, UDDG decode | query, count 1-10, timeFilter |
| scrapepage | Fetch + main-content heuristic + markdown | url, format=markdown|text|full, extractMainOnly |
| extractinsights | Entities, stats regex, key-point scoring, reading time | content, goal? |
| deepresearch | Power tool — search → parallel scrape → synthesize report | topic, depth=quick|standard|deep, maxSources, saveMemory |
| comparesources | 2-5 URLs → consensus vs unique vs contradictions | urls[], focus? |
| factcheckclaim | Searches support + debunked OR false, heuristic verdict | claim, searchDepth |
| memorysave | Save finding to JSON, survives restarts | key, value, tags[], source? |
| memorysearch | Fuzzy search in persistent memory | query, tags[], limit |
Resources:
research://memory — all saved findings
research://history — last 100 actions
research://stats — cache size, uptime
Prompts:
deep-dive-research — full research workflow
fact-check — fact-checker squad
compare-narratives — bias & comparison table
### Quickstart & Installation
#### ⚡ 1-Click Auto Setup for Claude Desktop (Recommended)
Run this single command in your terminal:
bash
npx mcp-deep-research-server --install
> That's it! It automatically detects your OS, writes the config to your claudedesktopconfig.json, and configures the 8 research tools. Just restart Claude Desktop.
#### 📦 Install via Smithery
To automatically install for Claude Desktop or Cursor via Smithery:
bash
npx -y @smithery/cli install mcp-deep-research-server --client claude
#### 🛠️ Manual / Other Clients Setup
Run via npx:
bash
npx mcp-deep-research-server
Or build from source:
bash
git clone https://github.com/SECRET4422/mcp-deep-research-server.git
cd mcp-deep-research-server
npm install
npm run build
#### 🤖 Agent Skill (Claude Code / Antigravity / Cursor / Agent Frameworks)
This repository includes an open-standard Agent Skill that teaches AI coding assistants when and how to autonomously run deep research workflows.
Import via skill directory: skills/deep-research/
Spec compliant with Claude Code, Cursor, and Google Antigravity agents.
### Test (smoke)
bash
npm run test:mcp
# or
npm run inspect # opens http://localhost:6274