The Agent Skills Directory

[SAFE]: The skill defines a controlled workflow for 'Contract-Driven Development' planning. It uses a read-only planning phase followed by an approval-gated apply step, ensuring that no modifications occur without user intervention. The skill configuration explicitly disables automatic model invocation (disable-model-invocation: true), which significantly reduces the risk of unauthorized tool use. No suspicious patterns, such as network exfiltration, hardcoded credentials, or obfuscated code, were detected in the instructions or configuration files.

cdd-plan