library-os
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, including OCR text from photos, Kindle exports, and voice memos, to generate structured book reviews.
- Ingestion points: Text extraction from user-provided photos and file reading of highlight exports.
- Boundary markers: The skill emphasizes a "curated" approach, implying human review of the structured output.
- Capability inventory: The workflow involves writing content to local TypeScript files (
data/book-reviews.ts) and performing Git operations. - Sanitization: While not detailed in this documentation, the manual curation step provides a verification layer.
- [COMMAND_EXECUTION]: The skill documents a workflow using Git commands to manage changes and deploy to the vendor's production infrastructure.
- Evidence: The 'Git workflow' section describes creating worktrees, committing changes to
app/library/, and pushing branches to thefrankxai/frankx.ai-vercel-websiterepository. - Note: These operations are consistent with the skill's primary purpose and target the vendor's own deployment environment.
Audit Metadata