The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill is designed to ingest and process several external markdown files (define/문제가설.md, define/문제검증인터뷰결과.md, etc.).
Ingestion points: The 'Required Inputs' section explicitly lists local markdown files as the source of truth for its analysis.
Boundary markers: None detected. The agent interpolates these files directly into its reasoning process.
Capability inventory: The skill primarily generates new markdown files (think/킹핀문제.md, think/문제해결방향성.md) and lacks high-privilege capabilities like network access or command execution.
Sanitization: Not present. The skill assumes trust in the local file system.
[No Code] (SAFE): The skill consists entirely of markdown instructions and templates. No executable scripts, binaries, or package dependencies are present, significantly reducing the attack surface.

direction-setting