amazon-data
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Data Exposure & Exfiltration] (SAFE): The skill handles authentication securely by instructing users to use environment variables for API keys rather than hardcoding them. It communicates only with the relevant Canopy API domain.
- [Indirect Prompt Injection] (SAFE): While the skill retrieves untrusted data from Amazon (reviews and product descriptions), it is used solely for data retrieval and lacks high-privilege capabilities like command execution or file system writing, mitigating the risk of injection attacks.
- [Unverifiable Dependencies] (SAFE): The skill references only the standard and widely trusted Python 'requests' library in its code examples.
Audit Metadata