amazon-pricing-command-center

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted metadata from Amazon product listings, which creates a surface for indirect prompt injection (Category 8).
  • Ingestion points: Product titles, features, and descriptions are fetched from the ZooData API via scripts/zoodata.py subcommands.
  • Boundary markers: Absent. The instructions do not define delimiters or provide warnings for the agent to ignore instructions embedded in product data.
  • Capability inventory: The skill is authorized to execute shell commands through the bundled scripts/zoodata.py script.
  • Sanitization: Absent. No explicit sanitization or filtering of external textual data is described in the skill.
  • [DATA_EXFILTRATION]: The bundled script scripts/zoodata.py reads from ~/.zoodata/config.json to retrieve the API key. While necessary for the tool's operation, this involves accessing a sensitive path in the user's home directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 11:17 AM
Security Audit — agent-trust-hub — amazon-pricing-command-center