analyze-video
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local shell script (
extract-frames.sh) and standard media processing binaries (ffmpeg,ffprobe) to extract data from user-provided video files. - [EXTERNAL_DOWNLOADS]: The skill instructions guide users to install reputable tools from well-known sources, including
ffmpegvia Homebrew and theopenai-whisperPython package via PyPI. - [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection through its ingestion of external video content and resulting audio transcripts.
- Ingestion points: Audio transcripts extracted from user-provided video files in Step 2.
- Boundary markers: No explicit delimiters are used to separate transcript data from processing instructions.
- Capability inventory: The skill has the ability to write new markdown files to the local prompt library and submit data to the Arcads API.
- Sanitization: No content validation or sanitization is performed on the extracted transcripts before analysis.
Audit Metadata