do-think
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or data exfiltration attempts were found in the skill's instructions or supporting framework files.
- [PROMPT_INJECTION]: The skill uses authoritative instructional language (e.g., 'Iron Laws') to enforce its internal reasoning process, but it does not attempt to bypass the system prompt's safety guidelines or core instructions.
- [DATA_EXFILTRATION]: No sensitive file access or unauthorized network operations were identified. The skill focuses on reasoning within the provided context.
- [REMOTE_CODE_EXECUTION]: The skill does not perform any external package installations or execute remote scripts. It uses a structured 'ask-user' pattern for interactive sessions, which is standard for agentic workflows.
- [INDIRECT_PROMPT_INJECTION]: While the skill is designed to process user-supplied problem statements, it does not possess dangerous capabilities like file writing or arbitrary command execution that would make it vulnerable to high-severity injection attacks. The risk remains low as it is limited to reasoning and planning.
Audit Metadata