CodeAct provides Think-Code-Observe agents that synthesize and execute Ruby code dynamically. Ship DSPy.rb workflows that write custom Ruby code while tracking execution history, observations, and safety signals.

DSPy datasets provide prebuilt loaders, caching, and schema metadata for benchmark corpora used in DSPy examples and teleprompters.

Planner, queue, and coherence orchestration built on DSPy::DeepSearch.

DeepSearch loop utilities and modules for DSPy agents.

Provides the DSPy::Evals runtime, concurrency, callbacks, and export helpers for benchmarking Ruby DSPy programs.

Ships DSPy::Teleprompt::GEPA plus reflective adapters, experiment tracking, and telemetry hooks built on top of the GEPA optimizer core gem.

Optional optimizer bundle for DSPy.rb that ships the MIPROv2 teleprompter, Gaussian Process backend, and supporting dependencies for Bayesian optimization.

Provides DSPy::Observability, AsyncSpanProcessor, and ObservationType so instrumentation can be enabled independently from the main DSPy gem.

GEPA delivers optimization strategies, telemetry, and proposer tooling for reflective DSPy agents.