dspy
Where dspy-agent-skills goes next
Three categories: next examples I'd add, production patterns the skills currently stop short of, and one speculative research idea I'd like to see someone try.
dspy
dspy
Inside the examples: how GEPA lifted a 1.2B model by 25 points
Three end-to-end runs on free OpenRouter models. Real baselines. Real optimized scores. The counterintuitive reason why a smaller task LM exercises GEPA better than a bigger one.
dspy
How to install and use dspy-agent-skills
Fifteen seconds to install. Five minutes to smoke-test. Twenty to forty minutes for a real GEPA run. Works in Claude Code, Codex CLI, and anything else that respects the Agent Skills spec.
dspy
Why I built dspy-agent-skills
DSPy turns prompts into programs. GEPA optimizes those programs without RL. RLM reasons over inputs bigger than any context window. dspy-agent-skills packages that workflow for coding agents.