The Agent Skills Directory

[SAFE]: The skill is designed for the narrow purpose of summarizing local plan files into structured text and Mermaid diagrams. It contains explicit instructions to avoid making inferences or accessing external data sources.
[SAFE]: The configuration includes disable-model-invocation: true, which restricts the agent's ability to call other tools or models during the execution of this specific skill, minimizing the attack surface.
[SAFE]: No indicators of malicious activity such as data exfiltration, prompt injection, or obfuscation were found in the skill's instructions or metadata.

plan-explain