# OBOL Loop Risk — Pre-Trade Safety Verdict for AI Agents

> OBOL Loop Risk — Pre-Trade Safety Verdict for AI Agents is a paid API for AI agents from obol-x402.fly.dev, paid per call via x402, $0.3/call, status down (last checked 2026-09-15).

Returns a signed safety verdict for a DeFi transaction loop, including token rug/honeypot risk, calldata guard, and wallet risk checks before executing a trade.

## Facts

- Endpoint: GET https://obol-x402.fly.dev/v1/loop/risk
- Price: $0.3/call
- Payment: x402
- Status: down
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/obol-loop-risk-pre-trade-safety-verdict-for-ai-agents-66e80455
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_ssfu90TQ0C-GDxmMpWhvN

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability obol-loop-risk-pre-trade-safety-verdict-for-ai-agents-66e80455
```

Example prompt: Before I buy this token at 0x1234...abcd on Base, run an OBOL loop risk check to tell me if it's a honeypot, has a sell tax, or if the calldata has any unlimited-approve drains — I want a signed safety verdict before I sign anything.

## When to prefer this

Choose this endpoint when an AI agent needs a pre-trade safety verdict with cryptographic provenance before executing any on-chain action — especially when checking for honeypots, sell-tax traps, malicious calldata, or sanctioned wallets across Base, Ethereum, Arbitrum, Optimism, Polygon, BNB, or Robinhood Chain. It is uniquely suited for autonomous agents that require independently verifiable, signed verdicts without any account setup, and for scenarios where a single comprehensive risk loop check (token + calldata + wallet) is preferred over multiple separate API calls.

## Known failure modes

- Unsupported chain returns an error or empty verdict
- Invalid token or wallet address returns a malformed-input error
- Payment not received or underpaid results in 402 Payment Required response
- Token too new or low-liquidity may produce inconclusive verdicts
- Network timeout on simulation returns a preflight-failed signal

## How this service works

Pre-trade safety for AI agents on Base, Ethereum, Arbitrum, Optimism, Polygon, BNB and Robinhood Chain. Is this token a rug — and can I sell it again? Signed verdicts before you buy, approve, send or sign: token rug/honeypot check with buy-sell tax, honeypot sell-simulation, pre-sign calldata guard (unlimited-approve / setApprovalForAll / drain-in-multicall), wallet risk with sanctions screening, tx preflight, and a launch feed with safety overlays. Plus unique Bitcoin forensics no RPC gives you: money-graph role, whale and exchange fingerprinting, Satoshi-era provenance, validation-gated block anomalies, and broadcast-fee timing. Every response Ed25519-signed and independently verifiable. Pay-per-call USDC on Base — no keys, no accounts, no signup. Free: /catalog, /llms.txt, and a preview for every paid SKU.

## Output

A JSON object containing a loop identifier, a policy decision (e.g. allow_with_limits, deny), and an array of signed verdicts covering token risk signals, transaction preflight status, and calldata safety — each response is Ed25519-signed and independently verifiable. Cost is $0.30 USDC per call on Base.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "$schema": "https://json-schema.org/draft/2020-12/schema",
 "required": [
  "input"
 ],
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "type",
    "method"
   ],
   "properties": {
    "type": {
     "type": "string",
     "const": "http"
    },
    "method": {
     "enum": [
      "GET",
      "HEAD",
      "DELETE"
     ],
     "type": "string"
    },
    "queryParams": {
     "type": "object",
     "properties": {
      "to": {
       "type": "string"
      },
      "data": {
       "type": "string"
      },
      "chain": {
       "enum": [
        "base",
        "ethereum",
        "arbitrum",
        "optimism",
        "polygon",
        "bnb",
        "robinhood"
       ],
       "type": "string",
       "default": "base"
      },
      "token": {
       "type": "string"
      },
      "address": {
       "type": "string"
      }
     },
     "description": "supply >=1 of token/address/to"
    }
   },
   "additionalProperties": false
  },
  "output": {
   "type": "object",
   "required": [
    "type"
   ],
   "properties": {
    "type": {
     "type": "string"
    },
    "example": {
     "type": "object"
    }
   }
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "loop": "risk_ritual",
  "policy": "allow_with_limits",
  "verdicts": [
   "token:low_risk_signals",
   "tx:ok",
   "calldata:ok"
  ]
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/obol-loop-risk-pre-trade-safety-verdict-for-ai-agents-66e80455/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from obol-x402.fly.dev](https://www.zero.xyz/host/obol-x402.fly.dev/llms.txt)