ContentAnalysis

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted external content from YouTube transcripts and web articles. Attacker-controlled instructions within that content could be interpreted by the agent during the extraction process.\n
  • Ingestion points: YouTube URLs, Article URLs, and text inputs in ExtractWisdom/Workflows/Extract.md.\n
  • Boundary markers: The workflow does not define delimiters or specific 'ignore' instructions for the external content.\n
  • Capability inventory: The skill executes shell commands via fabric and reads local user files.\n
  • Sanitization: No sanitization or validation of the fetched content is specified.\n- [EXTERNAL_DOWNLOADS]: The workflow involves fetching data from remote URLs to perform content analysis.\n- [COMMAND_EXECUTION]: The skill executes the fabric CLI tool via subprocess calls to retrieve YouTube transcripts and process content.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 11:02 PM