| name | blue-audit |
| description | AI security audit for Base smart contracts — critical issues, suggested fixes, go/no-go verdict. $1.00 x402. |
blue-audit — Smart Contract Security Review on Base
Performs AI-powered security and product risk review of Base smart contracts or project code. Returns: critical issues, severity ratings, suggested fixes, and a go/no-go verdict.
Trigger Conditions
Activate on:
- "Audit this contract" / "security review"
- "Is this safe to interact with"
- "Check this Solidity code / Base contract"
- "Rug check" / "honeypot check"
- Any request to review smart contract security before deploying or interacting
How to Call
POST https://blueagent.dev/api/x402/blue-audit
X-Payment: <EIP-3009 USDC, amount: 1000000 (1.00 USDC)>
Content-Type: application/json
{
"prompt": "Audit this contract: <contract address or paste Solidity code>"
}
Price: $1.00 USDC
Pay to: 0xb058a1e305d9c720aa5b1bf42b6f2f6294b03b5f
Manifest: https://blueagent.dev/.well-known/ai-tool/blue-audit.json
Output Structure
- VERDICT — GO / NO-GO / CONDITIONAL
- Critical issues — severity (CRITICAL / HIGH / MEDIUM / LOW), description, line reference if available
- Suggested fixes — concrete code changes per issue
- Risk summary — rug vectors, ownership risks, upgrade risks, liquidity risks
- What to verify onchain — Basescan checks, ownership renounce, liquidity lock
Key Rules
- All contract addresses must be Base Mainnet — flag if mainnet vs testnet unclear
- Never confirm a contract is "safe" — return "no critical issues found" with explicit caveats
- For address-only input: fetch from Basescan for analysis (note: bytecode-only = limited analysis)
- Distinguish between AI analysis and formal audit — always caveat that this is not a formal security audit
Output Modes
GO: No critical/high issues found. List mediums and lows with fix suggestions.
CONDITIONAL: High issues found with clear fix path. Detail what must change before deploying/interacting.
NO-GO: Critical issues. Clearly state what the vulnerability is and potential exploit scenario.
Integration
Pairs with: contract-trust (quick trust score), protocol-risk-monitor (ongoing risk tracking), blue-build (architecture review before audit)
