scientific-systematic-review
Pass
Audited by Gen Agent Trust Hub on Jul 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Analysis of the skill instructions and code shows no signs of malicious patterns, obfuscation, or unauthorized access to sensitive information.
- [COMMAND_EXECUTION]: The
generate_prisma_flowchartfunction inSKILL.mdincludes local file system operations usingos.makedirsand file writing. These operations are intended for artifact generation and do not target restricted or sensitive system paths. - [PROMPT_INJECTION]: The skill provides functions for processing scientific literature data. 1. Ingestion points:
screening_workflowinSKILL.mdingests literature records (titles/abstracts). 2. Boundary markers: Absent. 3. Capability inventory: Local file writing for diagram creation. 4. Sanitization: Absent. This represents a standard data processing attack surface for research tools and is not classified as a vulnerability in this context.
Audit Metadata