The Agent Skills Directory

[SAFE]: The skill's primary function is to process project context to generate vision statements. All file operations are restricted to the local documentation directory (defaulting to docs/project-overview/).
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it reads untrusted data from existing mission files or user input.
Ingestion points: docs/project-overview/mission.md and user-provided project context (SKILL.md, Execution Process Step 1).
Boundary markers: Absent. The instructions do not specify delimiters or instructions to ignore commands embedded within mission files.
Capability inventory: File-write capabilities to the docs/ directory.
Sanitization: Absent. The content from external files is used directly to inform the vision drafting process.
[EXTERNAL_DOWNLOADS]: The README.md mentions an installation command npx skills add nesnilnehc/ai-cortex. This refers to the author's own repository and is a standard procedure for adding skills in this environment.
[REMOTE_CODE_EXECUTION]: No patterns for remote code execution or unauthorized shell access were found in any of the analyzed files.

define-vision