ai-content-pipeline

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions and tools from the official inference.sh GitHub and domain. These are well-known resources for the purpose of the skill and do not present a security risk.
  • [COMMAND_EXECUTION]: The skill provides example CLI commands (belt app run ...) intended for the user to execute via the inference.sh command-line tool. These commands are descriptive of the skill's primary function (AI media generation) and do not contain malicious payloads.
  • [INDIRECT_PROMPT_INJECTION]: The skill has the surface area for processing user data (e.g., summarizing blog posts or scripts), but it is a documentation-centric skill providing templates rather than a runtime agent implementation that blindly executes untrusted input.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:18 PM
Security Audit — agent-trust-hub — ai-content-pipeline