{
  "id": "agentic-workflow",
  "code": "PTL-0064",
  "term": "Agentic Workflow",
  "aliases": [
    "LLM workflow"
  ],
  "category": "agents",
  "definition": "An agentic workflow is a system in which language models and tools are orchestrated through predefined code paths, as opposed to an agent that chooses its own steps.",
  "description": "Common workflow patterns include prompt chaining, routing, parallelization, orchestrator-workers, and evaluator-optimizer loops. Workflows are generally preferred when a task can be cleanly decomposed in advance.",
  "example": null,
  "broader": [],
  "narrower": [
    "prompt-chaining",
    "routing",
    "parallelization",
    "orchestrator-workers",
    "evaluator-optimizer"
  ],
  "related": [
    "ai-agent"
  ],
  "introduced": null,
  "sources": [
    {
      "title": "Building effective agents",
      "authors": "Anthropic",
      "year": 2024,
      "url": "https://www.anthropic.com/research/building-effective-agents"
    }
  ],
  "url": "https://protologue.com/t/agentic-workflow/",
  "citation": "Protologue. (2026). Agentic Workflow. In Protologue: A Taxonomy of Prompting and LLM Techniques (v1.0.0, PTL-0064). https://protologue.com/t/agentic-workflow/"
}