The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown instructions outlining a mental model for verification planning. It does not include any scripts, shell commands, or automated triggers.
[SAFE]: There are no network requests, external dependencies, or data exfiltration vectors present in the instructions.
[SAFE]: The methodology explicitly encourages safety and transparency, instructing the agent to keep verification support 'safe to inspect' and to 'ask before introducing dependencies'.
[SAFE]: No prompt injection or obfuscation techniques were detected in the text or metadata.

verification-planning