Compile declarative DSPy program signatures into optimized prompt and few-shot pipelines.
npx skills add https://github.com/ruiyangruiyi/hermes-agent --skill dspy-ruiyangruiyi
Please help me install this Agent Skill. Skill: dspy Source: https://github.com/ruiyangruiyi/hermes-agent/tree/main/skills/mlops/research/dspy Command: npx skills add https://github.com/ruiyangruiyi/hermes-agent --skill dspy-ruiyangruiyi
This Skill removes manual, trial-and-error prompt engineering by letting you define LM tasks declaratively and automatically optimize prompts using evaluation data.
Use it when you need more reliable outputs for QA, classification, extraction, or multi-stage RAG/agent systems, especially when you can supply labeled examples and a scoring metric.
Ask an AI to generate a DSPy module for question answering, then compile it with BootstrapFewShot using your labeled question-answer examples and an exact-match metric.
High-intent search queries and answers about installing and using this skill.
Automate prompt optimization by compiling declarative program signatures into self-improving pipelines using DSPy teleprompters, which automatically refine instruction text and few-shot examples based on a defined evaluation metric and training data.
Build a retrieval-augmented generation pipeline by composing modular reasoning blocks like ChainOfThought and ReAct, then apply automatic prompt optimization using teleprompters with a scoring metric over validation data to ensure consistent answer quality.
Declarative LM programming supports tool-using agents and structured extraction through modular workflows like ReAct and ProgramOfThought, allowing you to define input-to-output schemas and compile them for consistent execution quality.
You need to define input-to-output task signatures, configure an LM provider backend like OpenAI or Anthropic, and supply labeled train/validation data alongside a custom evaluation metric to run the teleprompter optimization successfully.
Automatic prompt optimization works with both OpenAI and Anthropic models, requiring you to configure the respective provider backend in your environment before executing the teleprompter compilation over your labeled dataset.
Avoid declarative LM programming when you lack labeled examples or a clear scoring metric, as teleprompters require train/validation data and an evaluation function to successfully compile and optimize the declarative program signatures.
ZardLi1115Automate declarative development and optimization of language model workflows.
zangjeicyCompile declarative language model calls into self-improving pipelines.
cryptopafiBuild AI pipelines with declarative signatures and automated prompt optimization.
SethyPagnaBuild declarative language model programs with structured signatures and optimization workflows.
