heuristics-and-checklists
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were identified across any of the evaluated files. The skill focuses on providing educational frameworks for decision-making and error prevention.
- [PROMPT_INJECTION]: The skill does not contain instructions that attempt to override agent safety protocols or bypass behavioral constraints.
- [DATA_EXFILTRATION]: There is no evidence of hardcoded credentials, sensitive file path access, or unauthorized network transmission of data.
- [REMOTE_CODE_EXECUTION]: The skill does not involve the installation of external packages or the execution of remote scripts.
- [OBFUSCATION]: Analysis did not detect any hidden content, multi-layer encoding, homoglyph attacks, or steganographic URLs within the text.
- [DYNAMIC_EXECUTION]: While the methodology discusses adaptive heuristics in pseudocode, the skill does not include any scripts that perform runtime compilation or dynamic code injection.
Audit Metadata