long-audio-to-obsidian

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script doc_merger.py and uses shell commands like mv to manage and organize files, which is consistent with its intended purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes transcription data, which serves as a surface for indirect prompt injection. 1. Ingestion points: Transcription files are read from the local directory by the doc_merger.py script. 2. Boundary markers: The script isolates content by wrapping it in Markdown code blocks. 3. Capability inventory: The skill can read, write, and move files within the project directory. 4. Sanitization: The skill relies on Markdown structure to separate external content from instructions without further sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 12:18 PM