react-expert
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data by reading repository files, specifically linting reports (
typescript.json,lint.json) and React component source code. This represents a vulnerability surface where malicious content embedded in the codebase could attempt to influence the agent's review logic. - Ingestion points: Reads project files and reports in
SKILL.md. - Boundary markers: None present in the instructions.
- Capability inventory: File reading and file writing within the project workspace.
- Sanitization: No explicit sanitization or input validation is defined for the processed content.
- [EXTERNAL_DOWNLOADS]: The README instructions reference fetching the skill from the author's GitHub repository and a repository belonging to a well-known organization. These downloads are part of the standard installation process for the skill platform and do not originate from untrusted sources.
- [COMMAND_EXECUTION]: The skill procedure requires the agent to read and write files within the project directory to perform its analysis and output the review report. These operations are essential to its primary purpose as a code review tool.
Audit Metadata