The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown instructions and reference documentation. It does not ship with any executable scripts, binaries, or configuration files that trigger automated installation procedures.
[PROMPT_INJECTION]: No malicious prompt injection patterns were detected. The instructions use natural language to define task boundaries and quality standards without attempting to override system safety filters or extract system prompts.
[DATA_EXFILTRATION]: The skill interacts exclusively with local project files in the .agents/ directory (e.g., root-cause.md, product-context.md). No network calls for data transmission or access to sensitive system paths (like ~/.ssh or environment variables) were identified.
[REMOTE_CODE_EXECUTION]: No patterns of remote code execution or external package dependencies were found. The skill does not perform any npm install, pip install, or curl | bash operations. The WebSearch directive is used as a standard information retrieval tool, not for executing remote code.

mkt-initiative-generate