Stats
Actions
Tags
From init
Routes users to appropriate dlt toolkits and skills for data engineering by listing catalogs, matching intents via skill descriptions, guiding installs, and confirming MCP setup.
How this skill is triggered — by the user, by Claude, or both
Slash command
/init:toolkit-dispatchThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Route the user to the right toolkit and skill.
Route the user to the right toolkit and skill.
Prefer MCP — use the list_toolkits tool from dlt-workspace-mcp to get the current toolkit catalog.
CLI fallback (if MCP is not connected): dlt --non-interactive ai toolkit list
Toolkits marked (installed: <version>) are ready to use. Others need installing first.
Use toolkit_info MCP tool (or dlt --non-interactive ai toolkit <name> info CLI) on each installed toolkit.
This returns skill names, descriptions (with "Use when..." patterns), and workflow rules — use these to match user intent.
Match the user's request to the best skill using descriptions from step 2. If no installed toolkit matches, suggest installing one.
Install command: dlt --non-interactive ai toolkit <name> install
uv run dlt ai status
npx claudepluginhub dlt-hub/dlthub-ai-workbench --plugin initDiscovers Claude Code tool environment including native tools and MCP servers via scans, amplifies prompts with capabilities, and suggests non-binding tool compositions for 'what tools to use' or 'best approach' queries.
Guides selection of spences10's Claude Code ecosystem tools including MCP servers, skills, CLIs via decision trees, recommended setups, and workflows.