Agentpmt No Account Agentaddress X402

SuspiciousAudited by ClawScan on May 13, 2026.

Overview

This payment skill is transparent about using AgentPMT/x402, but it lets an agent handle wallet private keys and make paid calls without clear built-in spending limits or approval gates.

Install only if you intentionally want the agent to make paid AgentPMT/x402 calls. Use a dedicated low-balance wallet, keep private keys out of prompts and logs, set explicit budgets and approval requirements, and verify every payee, network, token, amount, and dependency before use.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If the key is exposed, reused, or used by an over-permissive agent, credits or wallet funds could be spent or controlled by someone other than the user.

Why it was flagged

The instructions obtain a wallet private key from an external service and later use wallet keys for signing/payment flows. This is central to the stated purpose, but it is high-privilege financial credential handling.

Skill content
response = requests.post("https://www.agentpmt.com/api/external/agentaddress", timeout=30) ... private_key = wallet["evmPrivateKey"]
Recommendation

Use only a dedicated low-balance/disposable wallet or AgentAddress, keep keys in a real secret manager, do not paste keys into prompts or logs, and avoid sending unrelated funds to provider-generated addresses.

ConcernMedium Confidence
ASI02: Tool Misuse and Exploitation
What this means

An autonomous agent could make paid tool calls or spend wallet funds/credits in ways the user did not intend.

Why it was flagged

The x402 example selects the first payment option and builds a signed authorization for the returned payee and amount, without showing a user approval step or local spending limits.

Skill content
accepted = payment_required["accepts"][0] ... "to": accepted["payTo"].lower(), "value": str(accepted["amount"])
Recommendation

Require explicit user approval or a strict budget before signing payments; validate product/action slugs, payee, token, network, and maximum amount before each paid retry.

What this means

Installing packages in the active agent environment can introduce dependency risk if packages or versions are compromised.

Why it was flagged

The skill asks users to install unpinned Python dependencies. This is expected for HTTP requests and Ethereum signing, but dependency provenance and versions are not constrained.

Skill content
pip install requests eth-account
Recommendation

Install in an isolated environment and pin reviewed versions of requests and eth-account when possible.