ai-content-pipeline
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download the platform's CLI and related capabilities from the inference-sh organization on GitHub and via npx.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates multi-stage workflows where the output of one model (such as a generated script or image URL) is used as the input for a subsequent tool call. This is an inherent property of media pipelines.
- Ingestion points: Data generated from previous steps (scripts, URLs) is interpolated into inputs for tools like Kokoro TTS, Wan 2.5, and OmniHuman in
SKILL.mdexamples. - Boundary markers: None present in the example workflows.
- Capability inventory: The skill uses a Bash tool restricted to the
beltCLI command. - Sanitization: No explicit sanitization or validation of intermediate outputs is demonstrated in the examples.
Audit Metadata