# onchain.wick.pics — multi-chain token-safety for AI agents

> Token-safety intelligence for AI agents, bots & traders across PulseChain, Monad, Base & BSC — rug/honeypot verdicts (0–100 score + evidence), fresh-pool rug radar, size-aware exit-safety, and a wallet drainer-approval scanner. Deterministic on-chain analysis (no LLM in the verdict path — fast, cheap, reproducible). Notably covers **PulseChain**, which the major free tools (GoPlus, Honeypot.is) don't support at all. Free to start; pay-per-call via x402 (no signup, no key). Informational, not financial advice.

## Tools (MCP / API)
- check_token_safety — Is this token a scam? Verdict (SAFE … LIKELY_RUG) + 0–100 score + evidence.
- fresh_rug_radar — What just launched, and is it safe? Last 20 fresh pools, scored at creation.
- exit_safety — Could I sell $X at acceptable slippage? Size-aware price impact + verdict.
- check_ownership — Is ownership renounced/upgradeable, and what can an active owner still do (mint/blacklist/pause/tax)?
- safe_to_interact — One call → SAFE_TO_INTERACT / CAUTION / DO_NOT_INTERACT, bundled with reasons.
- wallet_approvals — A wallet's active ERC-20 approvals, flagging unlimited (drainer) grants.
- wallet_poison_check — Has this wallet been hit by address-poisoning? Flags lookalike / zero-value spoof addresses in its history before you reuse a "known" address.
- wallet_reputation — Should I transact WITH this wallet? Composite counterparty-hygiene verdict (CLEAN / LOW_RISK / CAUTION / HIGH_RISK): account shape, history depth, known-rug holdings, drainer-approval + poisoning exposure. Not an identity/KYC screen.
- ai_incident_lookup — AI-risk history for a company/vendor: 3.3K+ documented AI incidents, regulatory fines (EUR amounts), and SEC AI-risk disclosures. Vendor due-diligence before relying on an AI product.

## For agents
- MCP server: `npx -y @wickpics/onchain-safety-mcp` (stdio over the hosted API). Package: @wickpics/onchain-safety-mcp.
- x402 pay-per-call (USDC on Base, no signup): https://onchain.wick.pics/.well-known/x402.json
- OpenAPI 3.1 spec (auto-generate a client): https://onchain.wick.pics/openapi.json
- Agent authentication guide (auth.md standard): https://onchain.wick.pics/auth.md
- Agent integration docs: https://onchain.wick.pics/agents

## Chains
- PulseChain (chainId 369) — primary; NOT covered by GoPlus / Honeypot.is. If your agent touches PulseChain, this is the safety check.
- Monad, Base, BNB Smart Chain.

## Links
- Home: https://onchain.wick.pics/
- Buy API access (crypto, no signup/KYC): https://onchain.wick.pics/buy
- What Ticker (what alpha wallets are buying that also passes safety): https://onchain.wick.pics/what-ticker
- Agent-Token Reality Index (free): https://onchain.wick.pics/reality/ — the hyped AI-agent tokens ranked by "paper ratio" (real DEX liquidity ÷ market cap) + a contract-safety verdict each; machine-readable feed at /reality/data.json
- PulseChain Safety Leaderboard (free): https://onchain.wick.pics/pulse/ — the most-traded PulseChain tokens, each with a 0–100 safety verdict (the only safety board that covers PulseChain; machine-readable at /pulse/data.json).