Skill Detail
Iteratively optimize prompts and text-based agent configs against scored eval sets with GEPA
Use reflective search to improve prompts or text-configured agent components against a real eval set instead of manual prompt tweaking.
Templates & WorkflowsMulti-Framework
Templates & Workflows
Multi-Framework
Security Reviewed
⭐ 3.6k GitHub stars
⬇ 4.1M/wk npm
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill iteratively-optimize-prompts-and-text-based-agent-configs-against-scored-eval-sets-with-gepa
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Python environment, GEPA package, train and validation examples with a scoring function, model provider credentials or local models, target prompt or text configuration to optimize
Install & setup
Install GEPA with pip install gepa, prepare a scored train and validation set plus a seed candidate, then run the documented optimize flow or DSPy integration to generate and compare improved candidates.
Author
GEPA AI
Publisher
Organization
Last updated
Apr 16, 2026
Quick brief
Use GEPA when an agent builder needs to iteratively improve a prompt or other text-configured component against a scored eval set, not when they are just chatting with a model or browsing prompt tips. The workflow is bounded: provide a seed candidate, run reflective optimization against train and validation examples, review the improved candidate, and keep or reject it based on measured lift. That scope boundary, reflective text optimization against explicit evaluation data, makes this a skill rather than a generic model or framework listing.