{
  "id": "meta-prompting",
  "code": "PTL-0070",
  "term": "Meta-Prompting",
  "aliases": [],
  "category": "agents",
  "definition": "Meta-prompting uses a single model as a conductor that breaks a task down and writes prompts for fresh instances of itself acting as specialized experts, then integrates their outputs.",
  "description": "The term is also used more loosely for asking a model to write or improve a prompt.",
  "example": null,
  "broader": [],
  "narrower": [],
  "related": [
    "orchestrator-workers",
    "self-discover",
    "automatic-prompt-engineer"
  ],
  "introduced": 2024,
  "sources": [
    {
      "title": "Meta-Prompting: Enhancing Language Models with Task-Agnostic Scaffolding",
      "authors": "Suzgun & Kalai",
      "year": 2024,
      "url": "https://arxiv.org/abs/2401.12954"
    }
  ],
  "url": "https://protologue.com/t/meta-prompting/",
  "citation": "Protologue. (2026). Meta-Prompting. In Protologue: A Taxonomy of Prompting and LLM Techniques (v1.0.0, PTL-0070). https://protologue.com/t/meta-prompting/"
}