🧠 Build Think-Plan-Act AI Agents with Llama-4 Reasoning

This n8n workflow template implements a sophisticated 'Think-Plan-Act' architecture for building intelligent AI agents using Llama-4 via OpenRouter or similar providers. It transforms basic automations into smart agents that reason before acting: first in THINK mode, generating structured plans with goals, subgoals, tools, and assumptions; then in ACT mode, executing those plans step-by-step with tool integration and AI oversight.

Key features include built-in JSON parsers for reliable outputs, natural language fallbacks, modular loops for iterative reasoning, and a demo workflow (e.g., weather queries) for quick setup. Customize prompts, tools, and goals easily to fit any automation needs.

Benefits: Saves hours of development by providing a ready framework; enables tool-aware planning to avoid errors; composable for complex multi-agent systems. Ideal use cases: Enhancing CRM automations with decision-making, content generation pipelines, data analysis agents, or any task requiring dynamic planning like API orchestration or research bots.

Requirements: OpenRouter API key (compatible with OpenAI, Groq). Drop into existing flows for instant intelligence upgrade.