social-media-analyzer
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFE
Full Analysis
- PROMPT_INJECTION (SAFE): No behavioral overrides or safety bypass markers were detected in the skill instructions.
- DATA_EXFILTRATION (SAFE): No hardcoded credentials, sensitive file paths, or network operations were found.
- COMMAND_EXECUTION (SAFE): The skill uses local scripts for metric calculations; these scripts contain no dangerous functions like eval() or subprocess calls with untrusted input.
- INDIRECT_PROMPT_INJECTION (SAFE): [Ingestion points]: Campaign data in assets/sample_input.json. [Boundary markers]: None. [Capability inventory]: Local metrics calculation with no network or write-access. [Sanitization]: None. The finding is marked as SAFE as the data ingestion surface is essential for the skill's primary purpose and lacks executable capabilities.
Audit Metadata