issue-contract-writer
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses a defensive prompt engineering approach by requiring "finite delivery contracts" and explicit boundaries (In Scope/Exclusions), which naturally mitigates risks of open-ended execution or unintended behavior.
- [SAFE]: Included validation script
scripts/test_fixtures.pyperforms a self-audit using theast(Abstract Syntax Tree) module to verify that no forbidden or dynamic execution functions (e.g.,eval,exec,os.system) are present in the code. - [SAFE]: The skill maintains integrity through cryptographic hashing (SHA256) of its calibration fixtures, ensuring the test data has not been tampered with.
- [SAFE]: The skill explicitly restricts its own capabilities, stating it never creates or mutates GitHub issues, focusing solely on producing text and evidence.
Audit Metadata