The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions direct users to install the package using the npx skills command and provide setup scripts (install.sh, install.ps1) hosted on the developer's official GitHub repository.
[COMMAND_EXECUTION]: The skill orchestrates multiple sub-agents by invoking the Task (Cursor) or Agent (Claude Code) tools to perform parallel reasoning waves.
[COMMAND_EXECUTION]: Integrates with external research components (last30days and deep-research) to fetch live web signal and verify factual claims during the decision process.
[PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection by ingesting untrusted data from user questions and local repository files to formulate instructions for sub-agents.
Ingestion points: Local codebase context and user questions are gathered in a 'Step-0' frame.
Boundary markers: Uses a structured bulleted frame for context, though it lacks explicit delimiters or instructions to ignore embedded commands.
Capability inventory: Includes spawning autonomous sub-agents and performing web searches.
Sanitization: No automated sanitization or escaping of external context is performed before it is interpolated into sub-agent prompts.

council