The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown documentation and instructions. It does not include any scripts, binaries, or executable code.
[SAFE]: No network operations, file system modifications, or credential accesses are performed by the skill. It explicitly states in its safety notes that it is a pure reasoning protocol.
[SAFE]: The instructions are designed to increase the agent's skepticism and rigor, which generally improves safety by preventing the agent from making unverified or overconfident claims.

falsify