badchars/darknet-mcp-server

English | 简体中文 | 繁體中文 | 한국어 | Deutsch | Español | Français | Italiano | Dansk | 日本語 | Polski | Русский | Bosanski | العربية | Norsk | Português (Brasil) | ไทย | Türkçe | Українська | বাংলা | Ελληνικά | Tiếng Việt | हिन्दी

Dark web & threat intelligence for AI agents.

HIBP, ThreatFox, ransomware tracking, Tor .onion access, blockchain intel, exploit search, stealer logs, malware analysis — unified into a single MCP server. Your AI agent gets full-spectrum dark web intelligence on demand, not 16 browser tabs and manual correlation.

The Problem • How It’s Different • Quick Start • What The AI Can Do • Tools (66) • Data Sources • Architecture • Changelog • Contributing

---

The Problem

Dark web intelligence is the missing layer in every security investigation. Breach databases, ransomware trackers, Tor hidden services, malware sandboxes, stealer logs, blockchain forensics, exploit databases — the data you need is scattered across dozens of platforms, each with its own API, its own auth, its own rate limits, its own output format. Today you check HIBP in one tab, ThreatFox in another, browse ransomware leak sites through Tor, pull up MalwareBazaar for a hash, check blockchain transactions on a block explorer, and then spend an hour manually piecing it all together.

Traditional dark web intel workflow:
  check breach exposure           ->  HIBP web interface (paid API)
  search leaked credentials       ->  IntelligenceX web interface
  track ransomware groups         ->  ransomware.live + ransomlook.io (2 separate UIs)
  access .onion hidden services   ->  Tor Browser manually
  analyze malware samples         ->  Hybrid Analysis + MalwareBazaar (2 more UIs)
  check IP abuse history          ->  AbuseIPDB + GreyNoise (2 more UIs)
  trace cryptocurrency            ->  blockchain.info + ChainAbuse
  search for exploits             ->  Vulners web interface
  check phishing URLs             ->  PhishTank web interface
  correlate everything            ->  copy-paste into a report
  ────────────────────────────────
  Total: 60+ minutes per investigation, most of it switching contexts

darknet-mcp-server gives your AI agent 66 tools across 16 data sources via the Model Context Protocol. The agent queries all sources in parallel, correlates data across the surface and dark web, identifies threats, and presents a unified intelligence picture — in a single conversation.

With darknet-mcp-server:
  You: "Investigate the breach exposure and threat landscape for target.com"

  Agent: -> HIBP: 3 known breaches (Adobe 2013, LinkedIn 2021, Collection #1)
         -> ThreatFox: 2 IOCs associated with domain (C2 callback, phishing)
         -> URLhaus: 1 malicious URL hosted on subdomain
         -> Ransomware: No victim listings found (good)
         -> Stealer logs: 47 compromised employee credentials found
         -> OTX: 5 threat pulses referencing the domain
         -> AbuseIPDB: Primary IP has 12 abuse reports (brute force)
         -> "target.com has been in 3 data breaches exposing 2.1M records.
            47 employee credentials found in stealer logs — immediate
            password reset recommended. 2 active ThreatFox IOCs suggest
            ongoing targeting. No ransomware listings, but the abuse
            reports on the primary IP warrant investigation."

---

How It’s Different

Existing tools give you raw data one source at a time. darknet-mcp-server gives your AI agent the ability to reason across surface web and dark web intelligence simultaneously.

---

Quick Start

Option 1: npx (no install)

npx darknet-mcp-server

Free tools work immediately. No API keys required for ransomware tracking, breach listings, GreyNoise, blockchain, OTX, and more.

Option 2: Clone

git clone https://github.com/badchars/darknet-mcp-server.git
cd darknet-mcp-server
bun install

Environment variables (optional)

# Breach & credential intelligence
export HIBP_API_KEY=your-key           # Enables breach account search & paste search
export INTELX_API_KEY=your-key         # Enables 4 IntelligenceX tools

# Threat intelligence
export OTX_API_KEY=your-key            # Increases AlienVault OTX rate limits
export ABUSEIPDB_API_KEY=your-key      # Enables 4 AbuseIPDB tools
export ABUSECH_AUTH_KEY=your-key       # Higher rate limits for abuse.ch suite
export PULSEDIVE_API_KEY=your-key      # Higher rate limits for Pulsedive

# Stealer logs & credentials
export HUDSONROCK_API_KEY=your-key     # Enables 3 Hudson Rock stealer log tools

# Exploit & malware analysis
export VULNERS_API_KEY=your-key        # Enables Vulners search & exploit tools
export HYBRID_API_KEY=your-key         # Enables 3 Hybrid Analysis malware tools

# Phishing
export PHISHTANK_API_KEY=your-key      # Higher rate limits for PhishTank

# Tor SOCKS5 proxy (for .onion access)
export TOR_SOCKS_HOST=127.0.0.1       # Default: 127.0.0.1
export TOR_SOCKS_PORT=9050            # Default: 9050

All API keys are optional. Without them, you still get ransomware tracking, breach listings, GreyNoise, blockchain intelligence, OTX, Tor exit node checks, onion search, CIRCL onion lookup, and more.

Connect to your AI agent

Tor setup (optional, for .onion access)

The 7 Tor tools (tor_fetch_onion, tor_scrape_onion, etc.) require a running Tor SOCKS5 proxy.

# macOS
brew install tor && brew services start tor

# Linux
sudo apt install tor && sudo systemctl start tor

# Docker
docker run -d -p 9050:9050 dperson/torproxy

# Verify connectivity
npx darknet-mcp-server --check-tor

Start querying

You: "Check if example.com has been in any data breaches and look for stealer logs"

That’s it. The agent handles breach lookups, credential searches, threat correlation, and more automatically.

---

What The AI Can Do

Breach Investigation

You: "Check if example.com has been in any data breaches"

Agent: -> breachList {domain: "example.com"}
       -> breachSearch {account: "[email protected]"}
       -> stealer_domain {domain: "example.com"}
       -> intelx_search {term: "example.com"}
       -> "example.com appears in 5 known breaches:
          - ExampleDB 2023: 4.2M records (emails, passwords, IPs)
          - Collection #1: credential stuffing list
          - Combolist 2024: 890K email:password pairs
          Stealer logs: 23 compromised machines with saved credentials
          for example.com found in Hudson Rock database.
          IntelligenceX: 47 results across paste sites and dark web."

Ransomware Tracking

You: "What ransomware groups are targeting healthcare?"

Agent: -> ransomwareBySector {sector: "healthcare"}
       -> ransomwareRecent {}
       -> ransomlookRecent {}
       -> "Healthcare sector ransomware activity:
          142 victims across 12 groups in the last 12 months.
          Top groups: LockBit3 (38 victims), BlackCat/ALPHV (24),
          Cl0p (19), Royal (15).
          Most recent: MedCorp Inc. listed by LockBit3 (2 days ago).
          3 new healthcare victims in the last week across
          ransomware.live and RansomLook feeds."

Dark Web Recon

You: "Search for mentions of our company on .onion sites"

Agent: -> tor_search_onion {query: "AcmeCorp"}
       -> intelx_search {term: "acmecorp.com"}
       -> onion_lookup {address: "relevant-onion-address.onion"}
       -> "Dark web search results for AcmeCorp:
          Ahmia.fi: 3 .onion pages mention AcmeCorp
          - Forum post discussing AcmeCorp employee list (leak forum)
          - Marketplace listing for AcmeCorp database (unverified)
          - Blog post referencing AcmeCorp breach
          IntelligenceX: 12 results on paste sites
          Recommendation: investigate the forum post and marketplace
          listing for potential active data sales."

Malware Analysis

You: "Analyze this file hash for malware indicators: abc123def456..."

Agent: -> malware_search {hash: "abc123def456..."}
       -> malware_overview {sha256: "abc123def456..."}
       -> bazaarHash {hash: "abc123def456..."}
       -> threatfoxSearch {searchTerm: "abc123def456..."}
       -> otx_hash {hash: "abc123def456..."}
       -> "Malware analysis for abc123def456:
          Hybrid Analysis: MALICIOUS (threat score 95/100)
          - Family: AgentTesla (info stealer)
          - MITRE ATT&CK: T1555 (Credential Access), T1071 (C2)
          - Network IOCs: 3 C2 domains, 2 IPs
          MalwareBazaar: First seen 2024-03-15, tagged 'AgentTesla'
          ThreatFox: 2 IOC entries linking to same campaign
          OTX: Referenced in 4 threat pulses"

---

Tools Reference (66 tools)

---

CLI Usage

# List all available tools
npx darknet-mcp-server --list

# Check Tor SOCKS5 proxy connectivity
npx darknet-mcp-server --check-tor

# Run any tool directly
npx darknet-mcp-server --tool breachList '{"domain":"adobe.com"}'
npx darknet-mcp-server --tool ransomwareRecent '{}'
npx darknet-mcp-server --tool tor_search_onion '{"query":"marketplace"}'
npx darknet-mcp-server --tool btc_address '{"address":"1A1zP1eP5QGefi2DMPTfTL5SLmv7DivfNa"}'

# Tools requiring API keys
HIBP_API_KEY=your-key npx darknet-mcp-server --tool breachSearch '{"account":"[email protected]"}'
HYBRID_API_KEY=your-key npx darknet-mcp-server --tool malware_search '{"hash":"abc123..."}'

---

Data Sources (16)

---

Architecture

src/
  index.ts                # CLI entrypoint (--help, --list, --check-tor, stdio server)
  protocol/
    mcp-server.ts         # MCP server setup (stdio transport)
    tools.ts              # Tool registry — all 66 tools assembled here
  types/
    index.ts              # Shared types (ToolDef, ToolContext, ToolResult)
  utils/
    rate-limiter.ts       # Per-provider rate limiter
    cache.ts              # TTL cache for API responses
    tor-fetch.ts          # Tor SOCKS5 proxy HTTP client
    require-key.ts        # API key validation helper
  tor/                    # Tor Network tools (7)
  ransomware/             # Ransomware Intelligence tools (9)
  breach/                 # HIBP Breach tools (7)
  abusech/                # ThreatFox + URLhaus + MalwareBazaar tools (9)
  otx/                    # AlienVault OTX tools (5)
  abuseipdb/              # AbuseIPDB tools (4)
  greynoise/              # GreyNoise Community tools (2)
  pulsedive/              # Pulsedive tools (3)
  hudsonrock/             # Hudson Rock stealer log tools (3)
  vulners/                # Vulners exploit tools (3)
  blockchain/             # Blockchain Intelligence tools (4)
  hybrid/                 # Hybrid Analysis malware tools (3)
  onionlookup/            # CIRCL Onion Lookup tool (1)
  intelx/                 # IntelligenceX tools (4)
  phishing/               # PhishTank tool (1)
  meta/                   # Meta tools (1)

Design decisions:

• 16 providers, 1 server — Every data source is an independent module. The agent picks which tools to use based on the query.
• Per-provider rate limiters — Each data source has its own RateLimiter instance calibrated to that API’s limits. No shared bottleneck.
• TTL caching — Ransomware data (15min), breach lists (10min), abuse.ch (5min) results are cached to avoid redundant API calls during multi-tool workflows.
• Graceful degradation — Missing API keys don’t crash the server. Tools return descriptive error messages: “Set HIBP_API_KEY to enable breach account search.”
• DNS leak prevention — Tor .onion tools use socks5h:// protocol to resolve DNS through Tor, preventing DNS leaks to the local resolver.
• 4 dependencies — @modelcontextprotocol/sdk, zod, socks-proxy-agent, and cheerio. All clearnet HTTP via native fetch. All Tor traffic via SOCKS5.

---

Limitations

• HIBP account/paste search requires a paid API key ($3.50/month)
• IntelligenceX, AbuseIPDB, Hudson Rock, and Hybrid Analysis require API keys for their tools
• Tor .onion tools require a running Tor SOCKS5 proxy (not bundled)
• abuse.ch free tier has lower rate limits without ABUSECH_AUTH_KEY
• Ransomware.live and RansomLook data depends on upstream scraping frequency
• Blockchain tools support Bitcoin only (no Ethereum/Monero)
• PhishTank database can lag behind real-time phishing campaigns
• macOS / Linux tested (Windows not tested)

---

Part of the MCP Security Suite

---

For authorized security testing and assessment only. Always ensure you have proper authorization before performing intelligence gathering on any target.

MIT License • Built with Bun + TypeScript