The Agent Skills Directory

[SAFE]: The skill provides a structured methodology for product design and ideation, utilizing internal assets for process guidance.\n- [DATA_EXPOSURE_AND_EXFILTRATION]: Analysis of the skill instructions and asset files confirms the absence of hardcoded credentials, sensitive file access (e.g., .ssh, .aws), or network operations targeting untrusted domains.\n- [COMMAND_EXECUTION]: The skill does not contain instructions for arbitrary command execution or privilege escalation. It primarily interacts with the local filesystem to read and write design documentation.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data from design/problem.md and design/model.md. While this constitutes an attack surface for data processed at runtime, the risk is inherent to the skill's purpose as a design tool and does not indicate malicious intent. The instructions provide clear structural boundaries (e.g., 'Raw material for concepting' contract) for how this data should be used.\n- [REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of remote scripts. The mention of Figma plugin skills refers to standard tool integration for UI design.

product-concept