x402-payments

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Prompt Injection (SAFE): No malicious instructions or bypass attempts were found. The instructions are focused on defining a specific expert persona and providing guidance on payment protocols.
  • Data Exposure & Exfiltration (SAFE): No hardcoded credentials or sensitive file access patterns were detected. The skill proactively includes validation rules to prevent the hardcoding of secrets (macaroon secrets) and the logging of sensitive payment preimages.
  • Obfuscation (SAFE): No encoded strings, zero-width characters, or other forms of obfuscation were identified in the provided files.
  • External Dependencies & RCE (SAFE): The skill does not define any Python or Node.js dependencies and does not perform any remote code downloads or execution.
  • Indirect Prompt Injection (LOW): The skill reviews user-provided source code, which is an external data surface. 1. Ingestion points: Analyzes **/*.{ts,js} files as specified in validations.md. 2. Boundary markers: Absent; no specific delimiters are defined to isolate user code from instructions. 3. Capability inventory: No executable tools, network operations, or file-system writing capabilities are present. 4. Sanitization: No sanitization of code content is performed before processing.
  • Persistence and Privilege Escalation (SAFE): No commands or instructions related to system persistence or privilege modification were present.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:07 PM