legacy-cleanup
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes and modifies external codebases using powerful tools including
read_file,write_file,delete_file, andshell_execute. This creates a vulnerability surface where potentially malicious instructions embedded in the target codebase (e.g., in configuration files, comments, or manifests) could attempt to influence the agent's decision-making or execution flow. - [EXTERNAL_DOWNLOADS]: Referencing official Node.js documentation to provide context on package entry points and export patterns.
Audit Metadata