Composes modular DSPy pipelines for LM tasks with declarative AI.
npx skills add https://github.com/math-inc/OpenGauss --skill dspy-math-inc
Please help me install this Agent Skill. Skill: dspy Source: https://github.com/math-inc/OpenGauss/tree/main/skills/mlops/research/dspy Command: npx skills add https://github.com/math-inc/OpenGauss --skill dspy-math-inc
DSPy provides a framework to build complex AI systems as modular, declarative pipelines, enabling tooling for prompts, RAG, and agent workflows.
Install DSPy and run a simple Predict-based QA flow to see results.
High-intent search queries and answers about installing and using this skill.
Declarative RAG pipelines are built by composing DSPy modules like Predict, ChainOfThought, and ReAct, which enables modularity and reuse for multi-step reasoning tasks. This framework treats LM tasks as composable pipelines rather than hardcoded prompts.
Data-driven prompt optimization uses teleprompters like BootstrapFewShot, MIPRO, and COPRO to automatically tune and optimize prompts within declarative AI pipelines. This replaces manual prompt engineering with systematic, programmatic optimization workflows.
Yes, declarative signatures support safe module composition for multi-step reasoning and agent workflows. Modules like ReAct and ProgramOfThought provide flexible reasoning patterns that can be combined to build complex AI systems.
Structure agent workflows by composing declarative modules such as ChainOfThought and ReAct into modular pipelines, then apply teleprompter optimization. This approach enables reuse and data-driven prompt tuning instead of manual engineering.
The Skill lists no external dependencies, but requires a YAML frontmatter with name and description in SKILL.md. Optional references, scripts, and assets can be included to support the pipeline implementation.
Declarative AI pipelines may add unnecessary overhead for simple, single-step LM tasks that do not require modular composition, multi-step reasoning, or data-driven prompt optimization. For straightforward predictions without reuse needs, direct API calls may suffice.
OmidZamaniBuild and optimize RAG pipelines with ColBERTv2 in DSPy.
markwang2658Automate declarative prompt optimization for language model pipelines.
peteromalletBuild end-to-end LM pipelines with declarative DSPy components and optimizers.
guccangOptimize language model programs with declarative task definitions and automated prompt improvement.
