book-to-skill

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (MEDIUM): Privilege escalation risk in documentation (Category 5). The 'workflows/convert-formats.md' file includes 'sudo' commands for installing external utilities. Downgraded from HIGH to MEDIUM as it is part of the primary setup workflow and intended for user-mediated installation.
  • [PROMPT_INJECTION] (LOW): Vulnerability to Indirect Prompt Injection (Category 8). The skill processes untrusted input to generate executable instructions and persistent skill content. Evidence: (1) Ingestion points: Untrusted .txt book files processed in 'workflows/analyze-book.md' and 'workflows/convert-book.md'. (2) Boundary markers: Absent; no delimiters are used to wrap or isolate book content. (3) Capability inventory: File-system write operations to create new skills and directories. (4) Sanitization: None identified; extraction logic does not filter for embedded instructions in book text.
  • [EXTERNAL_DOWNLOADS] (LOW): External dependencies (Category 4). The skill relies on Calibre and Pandoc for format conversion. These are reputable tools but represent external code dependencies required for functionality.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:19 PM