The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill launches a local Node.js web server (scripts/server.js) via a shell script (scripts/start-server.sh) to provide a visual brainstorming companion. The server serves HTML content that is dynamically generated by the agent during the session.
[COMMAND_EXECUTION]: The skill executes shell scripts (scripts/start-server.sh and scripts/stop-server.sh) to manage the lifecycle of the local brainstorming server, including process backgrounding and termination based on owner PID or idle timeouts.
[INDIRECT_PROMPT_INJECTION]: The skill takes user design requirements and transforms them into visual mockups rendered in a browser, creating a surface for potential Cross-Site Scripting (XSS) if the user provides malicious payload strings.
Ingestion points: User input describing features, layouts, and requirements during the brainstorming dialogue (processed according to SKILL.md).
Boundary markers: The instructions guide the agent to use specific HTML fragments and frame templates but do not include explicit instructions to sanitize or ignore embedded script tags within user data.
Capability inventory: The agent has the capability to write HTML files to the local file system (SCREEN_DIR) and trigger a browser-based rendering via the local HTTP server.
Sanitization: No explicit sanitization or escaping mechanisms are defined in the instructions for user-provided strings before they are rendered in the visual companion.

brainstorming