skills/codagent-ai/agent-skills/init/Gen Agent Trust Hub

init

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs routine environment checks for the 'agent-validator' tool. It verifies the installed version is compatible (≥ 0.15) and ensures a local configuration file exists before proceeding.
  • [COMMAND_EXECUTION]: Executes 'agent-validator --version' to check for tool availability. This is a standard and safe practice for initialization scripts.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests the output of a local CLI command ('agent-validator --version'). This presents a minimal attack surface as it only extracts a version number for comparison and does not interpolate the output into executable logic or downstream prompts beyond simple string display.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 04:47 PM