Automate DSPy optimization with teleprompters, metrics, and training data.
npx skills add https://github.com/Qredence/skill-fleet --skill dspy-optimization
Please help me install this Agent Skill. Skill: dspy-optimization Source: https://github.com/Qredence/skill-fleet/tree/main/.skills/dspy-optimization Command: npx skills add https://github.com/Qredence/skill-fleet --skill dspy-optimization
DSPy optimization automates the tuning of DSPy programs by selecting optimal teleprompters, metrics, and training data to improve performance and reliability.
Provide a DSPy program, select an optimizer, and supply a training set to generate an optimized, tested program.
High-intent search queries and answers about installing and using this skill.
DSPy prompt optimization automates tuning by selecting teleprompters like BootstrapFewShot, KNNFewShot, MIPROv2, and GEPA to tailor demonstrations and improve program performance using training data.
Evaluating DSPy programs uses custom metrics and evaluation workflows to compare configurations, validate robustness, and measure performance across end-to-end optimization runs.
DSPy optimization supports applying MIPROv2 and GEPA independently or across workflows to tailor prompts, compile programs, and promote high-quality results.
End-to-end DSPy optimization workflows require a DSPy program, a selected optimizer, and a training set to compile, train, evaluate, and promote optimized programs.
BootstrapFewShot and related teleprompters require example training sets and the DSPy library to configure and run optimizations for generating tested programs.
Custom metrics in DSPy evaluation require correct configuration via repository scripts and references; improper metric setup limits robust validation and performance comparison across configurations.
OmidZamaniFine-tune a DSPy program into deployable model weights via BootstrapFinetune.
maruakshayIdentify and mitigate security risks in DSPy prompt programs and tool wrappers.
ruiyangruiyiCompile declarative DSPy program signatures into optimized prompt and few-shot pipelines.
icyzhCompile declarative language model calls into self-improving pipelines with automated optimization.
