The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches an output schema from the author's repository at https://raw.githubusercontent.com/csark0812/toolbox/main/.skeleton/references/output-schema.md. This is used to standardize the format of prototype results.
[DYNAMIC_EXECUTION]: The core functionality of the skill involves creating and running "minimal runnable artifacts" using various runtimes such as Node.js, Python, or Bun. This behavior is intended for the skill's primary purpose of validating design logic and UI concepts through live code execution.
[INDIRECT_PROMPT_INJECTION]: There is an inherent risk surface because the skill ingests user-supplied "design questions" and logic requirements, which are then used to generate the code that the agent executes. Maliciously crafted input could potentially influence the behavior of the generated prototype script.
Ingestion points: User-provided design questions and logic parameters processed in SKILL.md and LOGIC.md.
Boundary markers: None explicitly defined in the provided files to separate user input from the code generation prompt.
Capability inventory: Execution of arbitrary shell commands and scripts (Node.js, Python, Bun) as defined in LOGIC.md and UI.md.
Sanitization: No explicit sanitization or validation of user-supplied logic before it is incorporated into the runnable artifact.

prototype