ai-content-pipeline
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill leverages the
infshCLI through Bash to manage data flow between multiple AI models, which is necessary for the described media pipelines. It uses standard output redirection to JSON files for intermediate steps. - [EXTERNAL_DOWNLOADS]: Includes instructions to download official skill components from the
inference-shrepository using thenpx skillsutility, which is the standard mechanism for this ecosystem. - [SAFE]: All documented workflows use legitimate third-party services (OpenRouter, fal.ai, ByteDance, Google) and the author's own verified infrastructure (inference.sh). No signs of obfuscation, data exfiltration, or malicious intent were detected.
Audit Metadata