The Agent Skills Directory

[No Code] (SAFE): This skill is entirely instructional and template-based. It does not ship with any Python, Node.js, or shell scripts, which prevents the execution of malicious code.- [Indirect Prompt Injection] (LOW): The skill processes user-provided task descriptions to create structured plans. While it lacks explicit sanitization for instructions embedded in user input, the impact is low as the skill only generates static planning files. 1. Ingestion point: SKILL.md via [작업 설명] parameter. 2. Boundary markers: None specified. 3. Capability: Writing files to the plans/ directory. 4. Sanitization: None.

plan-creator