$ docker-agent toolsets TYPE SUMMARY a2a Connect to remote agents via the Agent-to-Agent (A2A) protocol api Create custom tools that call HTTP APIs background_agents Dispatch work to sub-agents concurrently and collect results background_jobs Run and manage long-running shell commands environment Report the OS and resolved shell (read-only, no arguments) fetch Read content from HTTP/HTTPS URLs file Read, write, and edit individual files filesystem Read, write, list, search, and navigate files and directories git Read-only git repository inspection: status, log, branches, show, blame lsp Connect to Language Server Protocol servers for code intelligence mcp Extend agents with external tools via the Model Context Protocol mcp_catalog Discover and activate remote MCP servers from the Docker MCP Catalog memory Persistent key-value storage for cross-session recall model_picker Let the agent pick between several models per turn open_url Open a fixed URL in the user's default browser openapi Generate tools automatically from an OpenAPI specification plan Shared persistent scratchpad for multi-agent collaboration rag Search document knowledge bases with hybrid retrieval (RAG) scheduler Schedule instructions to run at a time or on a recurring interval script Define custom shell scripts as named tools with typed parameters session_context Reference a previous session as context in the current one shell Execute shell commands in the user's environment tasks Persistent task database with priorities and dependencies think Step-by-step reasoning scratchpad for planning todo Manage a task list for complex multi-step workflows user_prompt Ask the user questions and collect interactive input webhook Send outbound notifications (Slack, Discord, Telegram, IFTTT, Teams, and more) [exit 0]