gemini

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the gemini command to perform analysis and grounding tasks, passing user-defined prompts and project paths to the CLI.
  • [EXTERNAL_DOWNLOADS]: Fetches the skill instructions from the author's public GitHub repository and references the official Google Gemini CLI repository for installation.
  • [PROMPT_INJECTION]: The skill processes untrusted data from the filesystem and the web, creating a surface for indirect prompt injection. 1. Ingestion points: Reads code from the project directory and fetches web content via search tools (SKILL.md). 2. Boundary markers: Uses structured Markdown headers to separate goal, context, and requirements. 3. Capability inventory: The gemini tool can access the filesystem, search the web, and execute tasks. 4. Sanitization: The skill does not implement explicit sanitization or instructions to ignore embedded commands in the processed data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 12:00 PM
Security Audit — agent-trust-hub — gemini