Deterministic Base contract admission control for agents. Decide whether a Base contract interaction should proceed before your agent buys, routes funds, approves, pays, or interacts.
$0.10/call via x402 · No API key neededCall before pay. Call before approve. Call before interact. Augur is deterministic preflight for Base contract actions.
Fastest path for integration: start with /skill.md if you want the shortest agent-oriented workflow, use /openapi.json for a formal schema, or install the local wrapper from /mcp.
Use this exact Base WETH request as the canonical integration check. It is the fastest path from evaluator traffic to one successful paid machine call.
Read /skill.md for the shortest setup path, or use OpenAPI if your client already consumes schemas.
Call /analyze?address=0x4200000000000000000000000000000000000006, let your x402 client handle the 402 payment challenge, and retry the same URL with PAYMENT-SIGNATURE.
curl -s "https://augurrisk.com/skill.md" curl -s "https://augurrisk.com/analyze?address=0x4200000000000000000000000000000000000006" \ -H "PAYMENT-SIGNATURE: <x402-payment-proof>" | jq
Expected first success: HTTP 200 JSON with decision: "allow", level: "safe", and score: 0. If the address is missing or malformed, Augur returns 422 before payment.
Augur is a deterministic contract admission gate for Base agents and the workflows around them. It gives you a fast first pass before you trust a contract.
Fetches on-chain bytecode for a Base mainnet contract address and runs 8 deterministic detectors to produce a default decision, policy recommendation, supporting findings, and a composite 0–100 score.
Scores are bytecode heuristics, not a full audit or guarantee. A safe result means no major bytecode-level risk signals were detected in this scan.
The value is clearest at decision points where an agent is about to trust or touch a contract.
Screen a token contract before a trading agent buys, quotes, or routes into it.
Screen a contract before a treasury, routing, or execution agent sends funds to it.
Screen a contract before an approval, listing, or other workflow treats it as acceptable.
V1 action-aware policy is intentionally narrow. It currently supports action=approve on Base and adds an action-level recommendation without replacing the base contract decision.
curl -s "https://augurrisk.com/analyze?address=0x4200000000000000000000000000000000000006&action=approve&spender=0x1111111111111111111111111111111111111111&chain=base" \ -H "PAYMENT-SIGNATURE: <x402-payment-proof>" | jq
In this path, the contract can stay top-level allow while the action-level result becomes warn. That lets a calling wallet or agent stay more cautious about the approval than about the contract in the abstract.
Current limit: action-aware V1 is only for approve. If no spender allowlist is configured, live requests log spender trust as unchecked and still return the narrower action-level policy.
curl -s "https://augurrisk.com/analyze?address=0x4200000000000000000000000000000000000006" \ -H "PAYMENT-SIGNATURE: <x402-payment-proof>" | jq
Pay with any x402-compatible client. Returns JSON with decision, recommended_policy, findings, score, level, and category_scores for a Base mainnet contract.
USDC · Settled via x402 protocol · No API key, no signup
x402 is an HTTP-native payment protocol - your agent pays per call automatically, no API key or signup needed.
These pages show where Augur fits into common agent workflows. Each one uses the same paid /analyze endpoint and returns the full screening response.
See exact Augur output on notable Base contracts before you wire the API into an agent policy.
Install the MCP wrapper directly with npx -y augurrisk-mcp, or use the links below for the broader machine-readable surface.