Crypto risk intelligence: 19 tools for rug pull detection, AI forensics, and token analysis.
73% of new crypto tokens are scams or honeypots. This MCP server lets your AI agent check any token before transacting β the cheapest insurance in crypto.
19 tools for rug pull detection, honeypot scoring, deployer tracking, holder deep-dive, KOL shill detection, social OSINT, and LLM forensic analysis (Claude Sonnet 4 / Opus 4).
No install needed. 1 free call/day, no auth:
curl -X POST https://cryptorugmunch.app/api/agent/v1/check-risk \
-H "Content-Type: application/json" \
-d '{"token_address": "7GCihgDB8fe6KNjn2MYtkzZcRjQy3t9GHdC8uHYmW2hr"}'
Response:
{
"token_address": "7GCihgDB8fe6KNjn2MYtkzZcRjQy3t9GHdC8uHYmW2hr",
"risk_score": 85,
"risk_level": "critical",
"recommendation": "AVOID",
"honeypot_risk": true,
"freeze_authority": true,
"token_name": "SuspiciousToken",
"token_symbol": "SCAM"
}
pip install rug-munch-mcp
Connect directly to our hosted server:
https://cryptorugmunch.app/mcp/sse
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"rug-munch": {
"command": "rug-munch-mcp",
"env": {}
}
}
}
{
"mcpServers": {
"rug-munch": {
"command": "rug-munch-mcp",
"env": {}
}
}
}
mcporter config add rug-munch --command "rug-munch-mcp"
mcporter call rug-munch.check_token_risk token_address=<MINT>
| Tool | What | Price |
|---|---|---|
check_token_risk | Quick risk score (0-100) + recommendation | Free (1/day) or $0.04 |
check_token_risk_premium | Full breakdown + deployer + social OSINT | $0.10 |
check_batch_risk | Up to 20 tokens at once | $0.30 |
check_deployer_history | Deployer wallet analysis β serial rugger? | $0.06 |
get_token_intelligence | Complete token profile (price, holders, LP) | $0.06 |
get_holder_deepdive | Top holders, concentration, sniper detection | $0.10 |
get_social_osint | Social media presence and red flags | $0.06 |
get_kol_shills | KOL/influencer shill detection | $0.06 |
get_coordinated_buys | Detect coordinated buying patterns | $0.04 |
check_blacklist | Community blacklist check | $0.02 |
check_scammer_wallet | Known scammer wallet check | $0.02 |
get_market_risk_index | Global market risk overview | $0.02 |
get_serial_ruggers | Known serial rugger wallets | $0.02 |
marcus_quick | AI verdict β Claude Sonnet 4 quick analysis | $0.15 |
marcus_forensics | AI forensics β full Claude Sonnet 4 investigation | $0.50 |
marcus_ultra | AI deep analysis β Claude Opus 4 | $2.00 |
marcus_thread | AI analysis thread for X/Twitter | $1.00 |
watch_token | 7-day webhook monitoring for risk changes | $0.20 |
get_api_status | Service health + accuracy metrics | Free |
check_token_risk β no auth needed# Get a key instantly (free, anonymous)
curl -X POST https://cryptorugmunch.app/api/agent/v1/keys/request \
-H "Content-Type: application/json" \
-d '{"email": "you@example.com"}'
# Use it on ALL 19 endpoints
curl -X POST https://cryptorugmunch.app/api/agent/v1/check-batch \
-H "Content-Type: application/json" \
-H "X-API-Key: mri_your_key_here" \
-d '{"tokens": ["token1", "token2", "token3"]}'
Pay per-call with USDC on Base or Solana. See x402 docs.
Compatible with x402-fetch, x402-axios, Coinbase AgentKit, and any x402-enabled client.
| Registry | Link |
|---|---|
| Official MCP Registry | io.github.amarodeabreu/rug-munch-mcp |
| PyPI | pip install rug-munch-mcp |
| Smithery | rug-munch-intelligence |
| Glama | rug-munch-intelligence |
| x402 Bazaar | cryptorugmunch.app |
| A2A Agent Card | /.well-known/agent.json |
| awesome-x402 | PR #28 merged |
# Service status & accuracy metrics
curl https://cryptorugmunch.app/api/agent/v1/status
# OpenAPI spec
curl https://cryptorugmunch.app/api/agent/v1/openapi.json
# Human-readable skill description
curl https://cryptorugmunch.app/api/agent/v1/skill.md
# x402 payment details
curl https://cryptorugmunch.app/.well-known/x402
# A2A agent card
curl https://cryptorugmunch.app/.well-known/agent.json
# MCP server info
curl https://cryptorugmunch.app/.well-known/mcp.json
Full Python example agent that uses x402 to check tokens before trading: β CryptoRugMunch/x402-trading-agent
Drop-in plugin for AgentKit-powered agents: β CryptoRugMunch/rug-agent-kit
Your Agent
β
βββ pip install rug-munch-mcp (local STDIO)
β βββ Talks to cryptorugmunch.app API
β
βββ SSE: cryptorugmunch.app/mcp/sse (remote, no install)
β
βββ Direct HTTP: cryptorugmunch.app/api/agent/v1/*
βββ Free tier (1/day, no auth)
βββ API key (1000/hr, X-API-Key header)
βββ x402 USDC (Payment-Signature header)
/api/agent/v1/statusPOST /api/agent/v1/feedback β report false positivesMIT
β οΈ Note: This project was formerly known as CryptoRugMunch. The $CRM token is not affiliated with Marcus Rug Intel. We operate on a Stars-based subscription model β no token required.
Built by Marcus Rug Intel β the cheapest insurance in crypto.
Source-derived launch command. Check the maintainerβs required arguments and credentials before running:
uvx rug-munch-mcpMerge this template into ~/Library/Application Support/Claude/claude_desktop_config.json. Keep existing servers. Add any arguments, credentials, and permissions required by the maintainer; this template has not been install-tested.
{
"mcpServers": {
"io-github-amarodeabreu-rug-munch-mcp": {
"command": "uvx",
"args": [
"rug-munch-mcp"
]
}
}
}Restart Claude Desktop completely for changes to take effect. Confirm the server appears connected in the clientβs tool list, then try a read-only example from its documentation.
Claude Desktop setup referencerug-munch-mcppypiio.github.amarodeabreu/rug-munch-mcp works with any MCP-compatible client. Copy the config snippet from the Configuration section above and add it to the file shown for your client, then restart the application.
~/Library/Application Support/Claude/claude_desktop_config.jsonRestart Claude Desktop completely for changes to take effect.~/.cursor/mcp.jsonRestart Cursor for changes to take effect..vscode/mcp.jsonReload VS Code window for changes to take effect.~/.codeium/windsurf/mcp_config.jsonRestart Windsurf for changes to take effect..mcp.jsonSave at the project root, then start Claude Code in that project and review the MCP server approval prompt. Keep real credentials out of shared files.