Connects to the ZARQ crypto risk API to run pre-trade safety checks and rating lookups for 198 tokens without requiring an API key. You get eight tools including crypto_safety_check for sub-100ms validation, crypto_rating for A+ to F trust scores with five-pillar breakdowns, and crypto_dtd for distance-to-default analysis with distress signals. Useful when you need collapse warnings or structural stress alerts before executing DeFi trades. The system claims 100% recall on token deaths in out-of-sample testing. Available as a remote SSE endpoint or local stdio server, both hitting the same free zarq.ai backend with daily updates.
Independent crypto risk intelligence via the Model Context Protocol (MCP). Trust Score ratings, Distance-to-Default analysis, structural collapse warnings, and pre-trade safety checks for 198 tokens.
No installation needed — connect directly:
{
"mcpServers": {
"zarq-crypto": {
"url": "https://mcp.zarq.ai/mcp"
}
}
}
pip install mcp httpx
python zarq_mcp_server.py
| Tool | Description |
|---|---|
crypto_safety_check | Quick pre-trade safety validation (<100ms) |
crypto_rating | Full Trust Score with 5-pillar breakdown (A+ to F) |
crypto_dtd | Distance-to-Default with 7 distress signals |
crypto_signals | Active Structural Collapse/Stress signals |
crypto_compare | Head-to-head token risk comparison |
crypto_distress_watch | All tokens with DtD < 2.0 |
crypto_alerts | Structural collapse/stress warnings |
crypto_ratings_bulk | Bulk ratings for multiple tokens |
Client (Claude, Cursor, etc.)
↓ MCP Protocol
ZARQ MCP Server
↓ HTTPS
zarq.ai API (free, no key required)
Free API at zarq.ai. No API key required during Early Access.
crypto risk defi safety trust-score crash-prediction distance-to-default ratings blockchain token-analysis
MIT
io.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage
io.github.mikerawsonnz/llm-orchestration-agent
io.github.mikerawsonnz/authenticated-llm-agent
labforgedev/copilot-memory-mcp
csoai-org/agent-prompt-injection-firewall-mcp
io.github.mikerawsonnz/authenticated-multi-llm-agent