# The trust layer for autonomous agents

> One call before any irreversible action. AI agents pay per call (USDC via x402, no account, no API key) for live safety verdicts: token rug-checks on the XRP Ledger today; address risk, transaction intent, and endpoint/tool trust next.

This is an MCP server. Agents pay per call in USDC via the x402 protocol
(no account, no API key). Connect over streamable HTTP.

## The suite (one question: “is this safe to do?”)
- [LIVE] Token safety — XRP Ledger — “Is this token going to rug me?”
- [IN DEVELOPMENT] Address & counterparty risk — “Is this address a scam, drainer, mixer, or sanctioned?”
- [PLANNED] Transaction intent — “If my agent signs this, what actually happens?”
- [PLANNED] Endpoint & tool trust — “Is this x402 endpoint or MCP tool safe to pay and call?”

## Live tools
- check_xrpl_token(issuer, currency): Live safety verdict for any XRP Ledger token: clawback and global-freeze hard rejects, issuer blackhole status, LP-burn fraction, pool depth, and a proprietary rug blocklist. [$0.005 / call] — https://cerberusindex.com/tools/check-xrpl-token.html
- check_issuer_reputation(issuer): Reputation for any XRP Ledger issuer: proprietary rug blocklist status plus historical detection outcomes from a production sniper pipeline. [$0.002 / call] — https://cerberusindex.com/tools/check-issuer-reputation.html
- get_new_pools(limit): Feed of newly created XRPL AMM pools with initial liquidity, trading fee, and safety-signal scores from launch-time screening. [$0.01 / call] — https://cerberusindex.com/tools/get-new-pools.html
- quote_amm(issuer, currency, xrp_in): Quote an XRP-to-token swap against a live XLS-30 AMM pool: tokens out, spot vs effective price, price impact, pool depth, and trading fee. [$0.001 / call] — https://cerberusindex.com/tools/quote-amm.html
- submit_feedback(message, tool, requested_capability): Free feedback channel for AI agents: request capabilities, report issues. High-volume requests get built first. [FREE] — https://cerberusindex.com/tools/submit-feedback.html

## Endpoints
- MCP (streamable-http): https://cerberusindex.com/mcp
- Full machine docs (one fetch): https://cerberusindex.com/llms-full.txt
- Markdown docs per tool: https://cerberusindex.com/tools/<slug>.md
- Agent card: https://cerberusindex.com/.well-known/agent.json
- Connect guide: https://cerberusindex.com/connect.html
- Live stats (JSON): https://cerberusindex.com/api/stats
- Accuracy scorecard (verdicts scored vs reality): https://cerberusindex.com/accuracy.html · JSON: https://cerberusindex.com/api/accuracy
- Rug Registry (confirmed rugs + dead-on-chain tokens): https://cerberusindex.com/registry.html · JSON: https://cerberusindex.com/api/registry

## Notes for agents
- Agents: use the free submit_feedback tool. High-volume requests get built first.
- Verdicts are informational risk signals with reasons, not financial advice:
  https://cerberusindex.com/disclaimer.html
- Terms: https://cerberusindex.com/terms.html · Privacy: https://cerberusindex.com/privacy.html