litreview
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill orchestrates the execution of multiple local Python scripts (
citation_tracker.py,framework_recommender.py,cross_search_aggregator.py) to handle session state, framework selection, and bibliometric aggregation. - [COMMAND_EXECUTION]: Detailed instructions are provided for the agent to perform low-level file manipulations, specifically unpacking DOCX files (ZIP archives) and modifying internal XML structures to resolve validation errors.
- [EXTERNAL_DOWNLOADS]: The skill requires the
docxNode.js package for document generation and references a validation script (scripts/office/validate.py) intended for runtime use that is not included in the skill's file list. - [SAFE]: The skill implements a 'Three-count audit' (searches/received/cited) as a security and integrity measure to prevent the model from hallucinating or padding research results with training data.
Audit Metadata