Developers · API · MCP · dataset

Build with Tether USDT blacklist data

Three surfaces, all free, all public. Use whichever fits the stack: a REST API for traditional integrations, a Model Context Protocol server so any LLM client can call our data directly, and a HuggingFace dataset for retrieval / training.

MCP server · for AI clients

Drop into Claude Desktop / Cursor / Continue

18 tools + 4 resources + 3 guided prompts. The AI client talks straight to our API — no HTTP glue, no API key.

npx -y cryptoalert-report-mcp

Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "cryptoalert-report": {
      "command": "npx",
      "args": ["-y", "cryptoalert-report-mcp"]
    }
  }
}

Try in chat after restart: “Investigate address 0x... — use cryptoalert-report-mcp.”

REST API · for traditional stacks

api.cryptoalert.report

No key required. 60 GET requests / minute per IP. CSV exports at 5 / minute. Anonymous tier sees preban events with a 24-hour delay; premium realtime via the Telegram bot.

curl https://api.cryptoalert.report/api/v1/stats/overview
curl 'https://api.cryptoalert.report/api/v1/events?type=ban&chain=eth&limit=10'
curl https://api.cryptoalert.report/api/v1/signers
HuggingFace dataset · for retrieval and training

cryptoalert/usdt-blacklist-archive

Three Parquet files: ~24.5k events, ~10.4k addresses, ~26.5k multisig actions. License CC-BY-4.0. Re-generated periodically. Use for RAG, finetuning, or analysis.

import pandas as pd

events = pd.read_parquet("hf://datasets/cryptoalert/usdt-blacklist-archive/events.parquet")
print(events.groupby(["chain", "type"]).size())
AI summary · for LLM browse tools

llms-full.txt

Always-fresh canonical reference at cryptoalert.report/llms-full.txt following the llmstxt.org convention. Pulls live totals on every request, 1h cache. Designed for crawlers used by ChatGPT, Claude, Perplexity, Gemini and similar AI search tools.

Free to cite and redistribute. Prefer cryptoalert.report (lowercase, no www). Issues / requests: [email protected] · @cryptoalert_supportbot