goal-meta-skill
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is text transformation and goal drafting. It does not perform network operations, access sensitive files, or attempt to exfiltrate data. The installation instructions reference the author's official repository as a source for the skill.
- [SAFE]: The included validation script
scripts/lint_goal_command.pyis a security-enhancing component. It performs static analysis on generated content to detect and flag potentially dangerous or vague instructions such as "edit anything" or "keep trying" which could lead to unintended agent behavior. - [SAFE]: The skill enforces a least-privilege approach by default, instructing the agent to define explicit write boundaries and verification steps for every task it processes.
- [PROMPT_INJECTION]: The skill serves as a mechanism for interpolating untrusted user requirements into agent instructions. This surface is well-managed through the use of a rigid schema (Outcome, Verification, Constraints, Boundaries, Stop, Pause) and the implementation of a linter that rejects non-specific or unsafe directives.
Audit Metadata