autumn-best-practices

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): The skill documentation recommends installing autumn-js and autumn-py from public registries. These are not from the trusted organization list. The severity is lowered to LOW as package installation is fundamental to the skill's primary purpose.
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface. The skill facilitates ingesting untrusted data from the Autumn API (e.g., via check and products.list methods) which influences billing and access logic. Evidence: (1) Ingestion points: Responses from api.useautumn.com processed in SKILL.md examples. (2) Boundary markers: Absent in provided snippets. (3) Capability inventory: Application access control and billing state management. (4) Sanitization: None mentioned in documentation examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:54 PM