The Agent Skills Directory

[SAFE]: The skill functions as a conversational guide for ideation and does not perform any high-risk operations or request sensitive permissions.\n- [COMMAND_EXECUTION]: There is no evidence of shell command execution, subprocess spawning, or system-level modifications within the skill instructions or scripts.\n- [DATA_EXFILTRATION]: The skill does not contain instructions for network operations or the access of sensitive files (e.g., credentials, SSH keys, or environment variables).\n- [DYNAMIC_EXECUTION]: The skill suggests generating temporary HTML and CSS for visual aids. The instructions explicitly state that these should be self-contained, low-fidelity, and disposable artifacts without external dependencies or complex logic, minimizing the risk associated with code generation.\n- [PROMPT_INJECTION]: No patterns were detected that attempt to override the agent's system instructions, bypass safety filters, or extract underlying prompts.\n- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided ideas, it lacks capabilities such as network access or file system writes, which significantly limits the potential impact of adversarial content embedded in user inputs.

brainstorm