The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown instructions and templates for organizing structured agent interactions. It does not include any scripts, executables, or dynamic code execution patterns.
[COMMAND_EXECUTION]: No shell commands, system-level operations, or privilege escalation patterns were detected. The skill operates entirely within the text-processing capabilities of the agent.
[DATA_EXPOSURE]: The skill references local project files such as design-state.md and design-memory for recording decisions. These are used for standard state persistence within the agent's expected ecosystem and do not involve sensitive system paths or hardcoded credentials.
[EXTERNAL_DOWNLOADS]: There are no network requests, external URL references, or package manager installations (NPM/Pip) identified in the content.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests context from design state files and user input, it lack dangerous capabilities (such as shell access or network calls) that could be exploited via malicious data. The risk is minimized by the structured templates provided for agent responses.

design-debate