agent-launcher-orchestrator
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates entirely within the local environment using standard Python libraries. Analysis of the scripts (
goal_router.py,goal_state.py,loop_compiler.py) confirms that no network requests are made and no external code is downloaded or executed.- [SAFE]: State management is handled through a local JSON file (goal.json) in a user-specified directory. The skill does not access sensitive system paths, credentials, or environment variables.- [SAFE]: The routing logic uses deterministic keyword matching rather than direct LLM interpretation of raw strings for control flow, which effectively mitigates common prompt injection and indirect instruction obedience risks.- [SAFE]: The skill adheres to best practices for security by generating deployment commands (e.g., curl) for the user to review and run manually, rather than handling API keys or performing automated launches.
Audit Metadata