gemini-analyzer

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • Command Execution (HIGH): The skill instructs the agent to format and execute shell commands by interpolating user-provided text into the Gemini CLI prompt argument (-p). This creates a significant risk of shell command injection if the user input contains metacharacters such as semicolons, pipes, or backticks.
  • External Downloads (SAFE): The skill utilizes the gemini-cli tool. According to the [TRUST-SCOPE-RULE], the google-gemini organization is a trusted source, so the dependency itself is considered safe.
  • Indirect Prompt Injection (LOW): The skill ingests untrusted data from the local codebase which could contain malicious instructions. Ingestion points: The --all-files flag causes the tool to read the entire codebase into the context. Boundary markers: There are no markers or warnings to prevent the agent from following instructions embedded in the codebase. Capability inventory: The skill can execute shell commands and read file contents. Sanitization: The skill does not perform any sanitization or escaping of the codebase content or the user's request parameters.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:37 PM