The Agent Skills Directory

[SAFE]: The skill provides instructions and code for building a monitoring dashboard for AI agents. The functionality is consistent with its stated purpose and provides useful developer tooling.
[EXTERNAL_DOWNLOADS]: Fetches standard, well-known Node.js packages (blessed, blessed-contrib, chalk, ws) from the official NPM registry for building the terminal interface and WebSocket communication layer. These dependencies are legitimate and commonly used for terminal dashboards.
[INDIRECT_PROMPT_INJECTION]: The dashboard processes and displays external data, such as tool inputs and outputs. The skill includes explicit security guidelines to respect privacy by truncating inputs and avoiding the logging of sensitive code content, which effectively manages the risks associated with processing untrusted data for display purposes.

claude-hud