dspy

Name: dspy
Availability: InStock
Author: samuelmukoti

Build modular AI pipelines with declarative DSPy programming.

Updated Apr 21, 2026

Search Tags:#RAG #agents #prompt-engineering #declarative-programming #dspy #modular-pipelines #LM-programming

Ready to run

npx skills add https://github.com/samuelmukoti/myai-agent --skill dspy-samuelmukoti

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: dspy
Source: https://github.com/samuelmukoti/myai-agent/tree/main/skills/mlops/research/dspy
Command: npx skills add https://github.com/samuelmukoti/myai-agent --skill dspy-samuelmukoti

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

DSPy provides a declarative programming model for building complex, modular AI pipelines that orchestrate LM-powered components (signatures, modules, and agents) with built-in optimization and tooling, reducing boilerplate and improving reliability.

Core Features & Use Cases

Declarative LM programming with signatures and modules to compose multi-stage systems (RAG, agents, classifiers).
Built-in optimization teleprompters (BootstrapFewShot, MIPRO, COPRO) to improve prompts and model behavior.
Type-safe, structured outputs with TypedPredictor for robust integrations and IDE support.
Supports multi-stage pipelines, tool-use agents, and cross-LM orchestration across providers.

Quick Start

Install DSPy, define a simple module, and run a basic prediction to see DSPy in action.

Frequently Asked Questions about dspy

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

What is declarative LM programming and how does it help build AI pipelines?▼

Declarative LM programming uses signatures and modules to compose multi-stage AI pipelines, reducing boilerplate and improving reliability. It orchestrates LM-powered components for systems like RAG, agents, and classifiers with built-in optimization.

How do I optimize prompts automatically for RAG systems and agents?▼

You optimize prompts for RAG systems and agents using built-in teleprompters like BootstrapFewShot, MIPRO, and COPRO. These optimizers automatically improve prompt behavior and model performance within modular DSPy pipelines.

Can I use multiple LM providers in a single multi-stage AI workflow?▼

Yes, you can use multiple LM providers in a single multi-stage AI workflow. The framework supports cross-LM orchestration across providers, allowing different stages of your pipeline to leverage different models.

Does this approach work for generating type-safe, structured outputs?▼

Yes, this approach works for generating type-safe, structured outputs. TypedPredictor provides structured outputs for robust integrations and IDE support, ensuring typed modules are exported for reuse across different components.

Do I need to install external libraries to build modular AI pipelines?▼

Yes, you need to install the DSPy core library to build modular AI pipelines. Optional teleprompters are also available for optimization if you want to automatically improve prompt behavior and model performance.