bro-think-with-me
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional markdown files and configuration data designed to guide an AI agent through various reasoning workflows without any executable scripts or binary components.
- [SAFE]: A robust READONLY policy is explicitly defined in the primary instructions (
SKILL.md) and reinforced in specific workflows (e.g.,debug.mdandexplorer-prompt.md), ensuring the agent does not attempt to modify the file system, databases, or system configuration. - [SAFE]: The code exploration capability is isolated using a subagent pattern with strict instructions to return only structured summaries rather than raw file content, which effectively mitigates the risk of processing malicious payloads embedded in analyzed code.
- [SAFE]: No evidence of obfuscation, sensitive data exfiltration, unauthorized network operations, or persistence mechanisms was found in any of the skill's components.
- [SAFE]: Metadata and model tier configurations use standard or placeholder identifiers that do not pose a security risk to the agent or its environment.
Audit Metadata