The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to create and run minimal code demos (referred to as 'spikes') in Python or TypeScript to verify technical assumptions. These operations are limited to the local project environment and are intended for standard development verification.
[DATA_EXFILTRATION]: The agent reads local project metadata and configuration from the .codestable/ directory to maintain context. No network exfiltration or access to sensitive user credentials (e.g., SSH, cloud keys) was detected.
[PROMPT_INJECTION]: Instructions define the agent's persona as a 'thinking partner' and provide a workflow for triaging user requests. These are operational guidelines and do not contain patterns aimed at bypassing safety filters or extracting system instructions.

cs-brainstorm