The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to perform an environment scan to identify installed companion tools.
Evidence: Executing ls on multiple skill directories (e.g., ~/.claude/skills/) within SKILL.md to detect missing recommended components.
Purpose: This check is used to offer context-aware installation commands for relevant complementary skills.
[EXTERNAL_DOWNLOADS]: Provides instructions for obtaining required SDKs and repositories from trusted public registries.
Evidence: Installation procedures for google-adk, @a2ui/web-lib, and flutter_genui via standard package managers.
Evidence: Clones the official project repository from https://github.com/google/a2ui.git for demonstration purposes.
[SAFE]: Demonstrates a focus on security by including explicit instructions for maintaining data privacy in multi-agent systems.
Evidence: Documentation in references/catalogs-and-actions.md advises stripping internal metadata before sharing data models to prevent cross-agent data leakage.

a2ui