The Agent Skills Directory

[SAFE]: The skill is primarily educational and architectural documentation. It provides code snippets and patterns for developers to follow when building terminal integrations. No malicious code or instructions were found.
[EXTERNAL_DOWNLOADS]: The documentation references standard industry libraries such as xterm, xterm-addon-fit, and xterm-addon-unicode11. It also links to GitHub Gists authored by GGPrompts for specific implementation examples (e.g., emoji width fixes, mouse coordinate transformations). These references are legitimate and serve the skill's primary purpose.
[COMMAND_EXECUTION]: The skill provides code examples demonstrating how to use the tmux CLI via execSync for session management operations like killing sessions or checking session health. These are standard practices for the described use case and do not represent a security risk within the context of the skill's guidance.
[DATA_EXFILTRATION]: No evidence of data exfiltration or hardcoded credentials was found. Network operations mentioned (like fetch calls to /api/tmux/detach/) are directed at local backend endpoints intended for session management.
[INDIRECT_PROMPT_INJECTION]: The skill describes systems that ingest untrusted terminal output and WebSocket messages. While this defines an attack surface (ingestion points in websocket-patterns.md and SKILL.md), the skill's purpose is to provide architectural guidance, and it appropriately inventories capabilities like execSync and PTY operations required for terminal multiplexing.

xterm-js