The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references numerous external software packages to be installed from the official NPM registry. These include well-known and trusted libraries for 3D rendering (Three.js), animation (GSAP, Framer Motion), and particle systems (tsparticles).
[COMMAND_EXECUTION]: The skill instructions (e.g., in references/kamehameha.md and references/over9000.md) direct the agent to use the shell (Bash) to install required technology stacks and libraries.
[PROMPT_INJECTION]: The references contain specific instructions labeled as 'Delegation Protocol' that steer the agent's operational behavior. Specifically, it instructs the agent not to use delegation tools (like the Task tool) but to perform all work directly using native file-editing and shell tools.
[DATA_EXPOSURE_AND_EXFILTRATION]: While the skill includes instructions for visual enhancements, it also describes an indirect prompt injection surface where the agent is instructed to read existing user components and apply modifications.
Ingestion points: The agent uses the Read tool to analyze existing project components (as noted in the Tool Coordination section of references/supersaiyan.md).
Boundary markers: The instructions do not define specific markers to delimit user-provided code from the agent's instructions.
Capability inventory: The skill encourages use of Edit/Write for code modification and Bash for package installation across all reference files.
Sanitization: There are no instructions for sanitizing or validating the content of the external code files before processing or modifying them.

visual-modes