setup-atome

Pass

Audited by Gen Agent Trust Hub on Nov 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill documents an integration that ingests data from external payment callbacks.
  • Ingestion points: Webhook callback endpoint (callbackUrl) in SKILL.md and atome-api.md receiving transaction reference identifiers.
  • Boundary markers: Instructions mandate the use of X-Signature HMAC-SHA256 verification and an authoritative server-side status requery.
  • Capability inventory: The skill performs network requests to the Atome API for payment creation and status checks.
  • Sanitization: The logic explicitly requires ignoring the callback payload's status and fetching the authoritative state from the payment provider via Basic Auth.
  • [EXTERNAL_DOWNLOADS]: The skill references downloads for platform plugins and SDKs from official vendor domains such as gateway.apaylater.com and vendor-associated S3 buckets.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 16, 2026, 05:14 AM