candid-friend
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional markdown files defining a reasoning and communication workflow. It does not contain any executable scripts, binary files, or automated commands that could pose a risk to the system environment.
- [SAFE]: The instructions emphasize user autonomy, directing the agent to always follow the user's final decision regardless of its own critique. This "remind and obey" logic is a safety-positive pattern that preserves human control.
- [SAFE]: References to external repositories, such as those on GitHub (e.g., garrytan/gstack), are provided solely for documentation and source attribution. These are well-known sources and do not involve automated code downloads or remote execution.
- [SAFE]: No obfuscation, prompt injection, or data exfiltration techniques were found in any of the skill files. The skill uses standard markdown and internal skill-to-skill references typical of modular agent platforms.
Audit Metadata