RiskOracle is NOT EP. NOT NoLeak. NOT MemGuard. Risk aggregation is its own layer — score before you act.
WHAT IT DOES
Autonomous agents fail in four distinct ways before any action: policy violation, wallet overexposure, memory drift, and execution leak. Four different failure modes. Four different services to check. RiskOracle calls all four and returns one normalized risk score with a recommended action — proceed, delay, or abort — and a cryptographic proof hash as an immutable record of every assessment.
API
POST /risk/check
Request:
{
"agentId": "your_agent",
"action": { "type": "swap", "capital": 50 },
"context": { "walletBalance": 500, "memoryState": {} }
}
Response:
{
"riskScore": 0.22,
"confidence": 0.91,
"breakdown": {
"policyViolation": 0.05,
"walletRisk": 0.08,
"memoryDrift": 0.03,
"executionLeak": 0.06
},
"recommendedAction": "proceed",
"proofHash": "0xa3f9c2e1...",
"latencyMs": 312
}
RISK THRESHOLDS
| riskScore | recommendedAction | Meaning |
| 0.00 – 0.30 | proceed | Safe to execute |
| 0.30 – 0.60 | delay | Elevated risk — wait or reduce size |
| 0.60 – 1.00 | abort | High risk — do not execute |
SCORING WEIGHTS
| Component | Weight | Source |
| policyViolation | 35% | EP /ep/validate |
| walletRisk | 30% | Internal wallet check |
| memoryDrift | 20% | MemGuard /memguard/check |
| executionLeak | 15% | NoLeak /noleak/check |
PRICING
| Tier | Price | Output |
| Basic | $0.005/call | riskScore only |
| Standard | $0.01/call | Full breakdown + confidence |
| Advanced | $0.02/call | recommendedAction + cross-check |
Also on Virtuals ACP marketplace — agent-to-agent.
AGENT LOOP POSITION
| RiskOracle | MemGuard | NoLeak | EP |
| Question | What is the overall risk? | Is my state valid? | Should I execute? | Is this authorized? |
| Endpoint | /risk/check | /memguard/check | /noleak/check | /ep/validate |
| Output | riskScore + action | driftScore + action | GER + confidence | proof_hash + policy |
| Connection | RiskOracle aggregates the other three — they remain fully independent |