clipcast-clipping

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a command-line interface clipcast and provides a fallback path to execute a local Python script located at /Users/prismosoft/Sites/Clipping/clipcast/main.py. This follows standard patterns for specialized CLI tooling.
  • [INDIRECT_PROMPT_INJECTION]: The tool ingests content from external video platforms (YouTube, TikTok, Instagram) via the --url parameter for transcription and highlight extraction. There is a potential for instructions embedded in video audio or transcripts to influence the AI's highlight selection logic.
  • Ingestion points: Video audio and metadata fetched from remote URLs provided to the --url flag.
  • Boundary markers: None specified in the user-facing CLI documentation.
  • Capability inventory: The tool performs audio transcription, file system writes to the outputs/ directory, and interaction with the Google Gemini API for content ranking.
  • Sanitization: The instruction set does not detail internal prompt sanitization for the transcription data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 09:03 PM