social-media-analyzer

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides marketing analysis functionality through documented workflows and mathematical calculations.
  • [COMMAND_EXECUTION]: The skill includes two Python scripts (scripts/calculate_metrics.py and scripts/analyze_performance.py) intended for data processing. A manual review of the source code confirms they only perform mathematical operations, sorting, and string generation. There are no calls to suspicious libraries, subprocesses, or network-capable modules.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill does not access sensitive local file paths (e.g., SSH keys, credentials) or environment variables. It operates entirely on the provided campaign data.
  • [PROMPT_INJECTION]: The instructions in SKILL.md and HOW_TO_USE.md provide structured guidelines for campaign evaluation and do not contain any patterns designed to override agent safety or system prompts.
  • [EXTERNAL_DOWNLOADS]: No external dependencies, remote scripts, or third-party package installations are required or performed by this skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 06:56 AM
Security Audit — agent-trust-hub — social-media-analyzer