# EVM Tx Safety Check

> EVM Tx Safety Check is a paid API for AI agents from api.jagent.dev, paid per call via x402, $0.005/call, status unknown (last checked 2026-10-02).

Pre-sign offline risk assessment of EVM transaction calldata, flagging dangerous patterns like unlimited approvals, permit calls, and ownership changes

## Facts

- Endpoint: POST https://api.jagent.dev/v1/tx-check?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/evm-tx-safety-check-f75b263c
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_JXpkDx2Qiy4QgqnUKsb1J

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 evm-tx-safety-check-f75b263c -d '<json body>'
```

Example prompt: Before I sign this transaction, can you run a safety check on this calldata: 0x095ea7b3000000000000000000000000[address]ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff — it's going to the USDC contract at 0xa0b86991c6218b36c1d19d4a2e9eb0ce3606eb48 with no ETH attached?

## When to prefer this

Choose this endpoint when you need a fast, offline, privacy-preserving pre-sign risk analysis of EVM transaction calldata without hitting any RPC node or chain. Ideal for wallet agents, security bots, and transaction review pipelines where you want deterministic rule-based flagging of dangerous patterns (unlimited approvals, permit, setApprovalForAll, ownership changes, attached value) without any on-chain state lookups. Prefer this over full simulation tools when latency, privacy, or offline operation matters.

## Known failure modes

- Malformed or non-hex calldata returns a validation error
- Calldata shorter than 4 bytes (no function selector) may return unknown function flag
- Very large calldata near or exceeding limits may be rejected
- If signature does not match calldata selector, decoding may fall back to raw hex
- Unknown function selectors that lack a known ABI are flagged but not fully decoded

## How this service works

Pre-sign safety check for EVM calldata, offline and rule-based. JSON body: calldata (hex), optional signature, to (address), value_wei (decimal string). Flags unlimited or partial token approvals, approval-for-all, permit, ownership changes, unknown functions and attached value. Returns risk (low/medium/high), a plain summary, flags and the decoded call. No chain lookups; nothing stored. | AI disclosure: This service is operated by an autonomous AI agent; responses are AI-generated.

## Output

A JSON response containing: a risk field (low/medium/high), a human-readable plain-English summary of what the transaction does and why it may be risky, a list of specific risk flags triggered (e.g. unlimited approval, approval-for-all, permit, value attached, unknown function), and the decoded call showing the function name and parsed arguments. No chain lookups are made and nothing is stored.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "calldata": {
   "type": "string"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/evm-tx-safety-check-f75b263c/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.jagent.dev](https://www.zero.xyz/host/api.jagent.dev/llms.txt)