adk-agent-builder

Scaffold a new Google ADK 2.0 agent (LlmAgent, workflow, custom, or no-code config)

Use this skill to define an ADK 2.0 agent declaratively in YAML/JSON via Agent Config — no Python code required. Triggers on: "ADK agent config", "no-code ADK agent", "agent.yaml", "declare ADK agent in YAML", "build agent without code", "ADK config file". Generates an agent.yaml describing model, instruction, tools, and sub_agents that ADK loads at runtime. Best for non-engineers, quick iteration, or storing agent definitions in a config repo.

Use this skill to subclass google.adk.agents.BaseAgent and implement a custom agent with full control over the run loop, event emission, and state transitions. Triggers on: "custom ADK agent", "subclass BaseAgent", "extend ADK with my own agent class", "non-LLM agent in ADK", "deterministic agent with custom logic", "ADK agent without an LLM call". Generates a class that overrides _run_async_impl and yields Events, ready to drop into a workflow or coordinator hierarchy.

Use this skill when the user wants to create a new ADK 2.0 LlmAgent or Agent in Python. Triggers on: "scaffold an ADK agent", "create a Gemini agent", "new LlmAgent", "build an ADK agent that uses tools", "google-adk agent starter", "set up a root_agent". Generates a runnable agent.py with google.adk.agents.Agent or LlmAgent, model selection (gemini-2.5-flash by default), instruction, description, and a tools=[] list ready for binding. Do NOT use for multi-agent coordinator setups (use coordinator-subagent-pattern) or workflow agents (use workflow-agent-scaffold).

Use this skill to wire ADK 2.0 agents that route across multiple LLM providers — Gemini, Gemma, Anthropic Claude, Ollama, vLLM, LiteLLM, Apigee AI Gateway, LiteRT-LM. Triggers on: "use Claude in ADK", "swap Gemini for Llama", "ADK with Ollama", "ADK LiteLLM", "ADK multi-provider", "route between models in ADK", "fallback model in ADK". Generates model-string config or a custom model adapter that ADK passes to the agent's model parameter, plus optional cost/latency-aware routing logic.

Use this skill to scaffold an ADK 2.0 Workflow Agent — Sequential, Loop, or Parallel — for deterministic multi-step pipelines. Triggers on: "ADK sequential agent", "build an ADK pipeline", "loop agent that retries", "run agents in parallel", "ADK workflow agent", "chain ADK agents", "deterministic agent flow". Generates SequentialAgent / LoopAgent / ParallelAgent code wiring N sub-LlmAgents in the requested topology, with termination conditions for loops. Use when control flow is fixed and deterministic — for code-driven branching/iteration use dynamic-workflow-builder, for graph-routed workflows use graph-workflow-designer.