content-repurposing

Pass

Audited by Gen Agent Trust Hub on Apr 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill utilizes the infsh CLI tool for its primary functions. The tool's execution is scoped via the frontmatter configuration, and its usage is consistent with the skill's stated purpose of content conversion.\n- [SAFE]: External dependencies recommended for installation via npx point to the vendor's repository (inference-sh/skills), ensuring a consistent trust chain.\n- [SAFE]: The skill references installation documentation hosted on GitHub, which is a standard and safe practice for distributing developer tools.\n- [PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by processing external media (e.g., podcast audio files) for transcription and summarization.\n
  • Ingestion points: Audio and text files ingested via the infsh app modules.\n
  • Boundary markers: Not explicitly defined in the provided command templates.\n
  • Capability inventory: Subprocess execution is restricted to the infsh tool; can perform write operations to social media via specific app modules.\n
  • Sanitization: Not explicitly present in the instructional templates, relying on model-level safeguards.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 23, 2026, 06:43 AM
Security Audit — agent-trust-hub — content-repurposing