setup-antom

Pass

Audited by Gen Agent Trust Hub on Nov 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill primarily consists of technical documentation and integration guides for the Antom payment service. It correctly emphasizes the use of placeholders for sensitive credentials and instructs developers to keep private keys on the server-side to prevent credential exposure.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents the handling of external payment notifications (webhooks), which represent an untrusted input surface. This risk is mitigated through robust security instructions documented in the reference files.
  • Ingestion points: Webhook notification endpoints defined in 'references/antom-notifications.md'.
  • Boundary markers: The skill mandates RSA256 signature verification over the raw body as a security boundary that must be passed before the agent or application processes any notification data.
  • Capability inventory: Instructions involve calling official Antom APIs, cryptographic signature operations, and idempotent database updates.
  • Sanitization: The instructions specifically advise verifying the signature against the raw body bytes rather than parsed JSON to prevent tampering via re-serialization.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 16, 2026, 05:14 AM