cross-disciplinary-bridge-finder
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implementation in
scripts/main.pyperforms local graph analysis using a predefined knowledge base of scientific domains. It does not perform any network operations or access sensitive system resources. - [COMMAND_EXECUTION]: The skill utilizes standard command-line arguments to accept domain names and configuration parameters. It includes functionality to save results to a local file, which is consistent with its stated purpose as an evidence insight tool.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface. 1. Ingestion points: Command line arguments
--source,--target, and--domainsinscripts/main.py. 2. Boundary markers: None. 3. Capability inventory: File write via the--saveparameter. 4. Sanitization: None. The risk is minimized as the inputs are used for dictionary-based lookups and formatting within a structured report.
Audit Metadata