review-fix-gemini
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches and runs the
@willbooster/agent-skills@latestpackage viabunx. This package is a vendor-owned resource from the skill author used to perform the review process. - [COMMAND_EXECUTION]: Executes shell commands to run the review tool and manage Git operations, including committing and pushing code changes to the repository.
- [PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection by processing external data as instructions.
- Ingestion points: The agent ingests output from the
reviewcommand as candidate comments for code modification (SKILL.md, Step 2). - Boundary markers: Absent. The skill does not implement delimiters or provide instructions to the agent to disregard commands embedded within the review tool output.
- Capability inventory: The agent possesses the capability to modify any file within the project and push those changes to a remote repository (Step 4 & 5).
- Sanitization: Absent. No logic is present to validate, escape, or filter the content of the review comments before the agent applies them to the codebase.
Audit Metadata