gemini
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the use of the
geminiCLI and includes troubleshooting steps involving process management commands such asps,lsof,pkill, andkill. These are used to detect and terminate hung instances of the model execution. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze large external contexts, such as codebases and technical documentation, which constitutes a potential attack surface for indirect prompt injection.
- Ingestion points: The agent reads user-provided codebases and documentation folders for analysis as described in the README.md and SKILL.md.
- Boundary markers: There are no explicit instructions or delimiters defined to isolate model instructions from the data being analyzed.
- Capability inventory: The
geminiCLI supports model-driven tool execution; when run with the--approval-mode yoloflag, the model can execute these tools without manual user intervention. - Sanitization: The skill does not implement or mention sanitization, validation, or filtering of the external codebase content before processing it via the Gemini model.
Audit Metadata