pay-with-app

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell commands including curl for API requests, jq for JSON processing, and cast (Foundry) for blockchain RPC interactions. It also executes dynamic Node.js scripts to generate cryptographic signatures.
  • [EXTERNAL_DOWNLOADS]: The skill downloads the viem library from the npm registry if it is not already present in the user's environment. This is handled with a specific security protocol that requires user consent via AskUserQuestion before installation.
  • [PROMPT_INJECTION]: The skill is designed to process external, untrusted data from HTTP 402 response bodies. It mitigates indirect prompt injection risks by enforcing strict validation rules, such as regex checks for addresses and amounts, and a blocklist for shell metacharacters in free-text fields.
  • [CREDENTIALS_UNSAFE]: The skill relies on sensitive environment variables (PRIVATE_KEY and UNISWAP_API_KEY). It includes explicit instructions to the agent to never hardcode or commit these secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 03:34 AM