using-gabrielmoreira-skills

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill provides explicit instructions to prevent data exfiltration, specifically forbidding the commitment of credentials, environment values, machine paths, or company names to version control.
  • [DYNAMIC_EXECUTION]: The skill instructs the agent to perform throwaway analysis by generating and executing Node.js scripts (.mjs) within a repository-root folder named '.local/'. This is a standard workflow for coding agents and is used here for data analysis rather than system modification.
  • [COMMAND_EXECUTION]: The instructions specify using the 'node' command to run generated analysis scripts and suggest using subagents or scripts for long-running shell tasks to maintain context efficiency.
  • [SAFE]: The 'evals/activation.scenarios.mjs' file contains only static test data for evaluating when the skill should be triggered and does not contain any executable logic or external network calls.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 04:17 PM
Security Audit — agent-trust-hub — using-gabrielmoreira-skills