Prompt Chaining
Also called LLM chains, multi-step prompting.
Prompt chaining decomposes a task into a fixed sequence of model calls, where each call processes the output of the previous one, often with programmatic checks between steps.
Description
Chaining trades latency for accuracy by making each call simpler. Wu et al. found chaining also improved transparency and controllability for users building with models.
Sources
- Wu et al. (2021). AI Chains: Transparent and Controllable Human-AI Interaction by Chaining Large Language Model Prompts.
- Anthropic (2024). Building effective agents.
Cite this entry
Protologue. (2026). Prompt Chaining. In Protologue: A Taxonomy of Prompting and LLM Techniques (v1.0.0, PTL-0065). https://protologue.com/t/prompt-chaining/
BibTeX
@misc{protologue_prompt_chaining,
title = {Prompt Chaining},
author = {{Protologue}},
year = {2026},
howpublished = {Protologue: A Taxonomy of Prompting and LLM Techniques, v1.0.0},
note = {Entry PTL-0065},
url = {https://protologue.com/t/prompt-chaining/}
}