outcome-bounded-work
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust framework for auditing instructions and shaping conversations. Its primary purpose is to distinguish between 'outcome invariants' and 'candidate paths' to ensure that requirements remain fixed while implementations remain flexible.
- [SAFE]: The skill includes several Python scripts (
scripts/check_code_rules.py,scripts/check_evals.py,scripts/lint_writing.py,scripts/validate_skill.py) used for static analysis, linting, and validation of the skill package itself. These scripts only perform local file operations and standard string/AST processing; they do not perform network operations, access sensitive system paths, or execute untrusted code. - [SAFE]: The skill uses a
mise.tomlfile to define a task graph for local development tasks (e.g., linting, testing, validation). These tasks trigger the included local scripts and do not involve remote code execution or external downloads. - [SAFE]: Instructions in
SKILL.mdemphasize safety and user authorization, specifically prohibiting the mutation of source artifacts without explicit user approval and the presence of a 'rightful editing owner' skill.
Audit Metadata