ai-social-media-content

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references external resources, including installation instructions from a GitHub repository and supplementary skills from the provider's registry.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided inputs such as content concepts and topic lists, which are interpolated into command arguments, presenting a standard attack surface for instructions embedded in data.
  • Ingestion points: The skill uses variable interpolation in shell scripts (e.g., $CONCEPT, $topic) within SKILL.md to pass user-defined data to model execution commands.
  • Boundary markers: No explicit delimiters or instructions are used to distinguish between data and commands for the backend models.
  • Capability inventory: The skill utilizes Bash(belt *) for command execution and local file redirection to save results.
  • Sanitization: No sanitization or escaping mechanisms are present for the interpolated variables.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:54 PM
Security Audit — agent-trust-hub — ai-social-media-content