Dspy Gepa Optimizer

Optimize DSPy programs with dspy.GEPA — a reflective/evolutionary optimizer to consider against task-specific baselines within an authorized evaluation budget. Use when the user says optimize, compile, GEPA, reflective optimization, or "make this program better" and a DSPy program + metric + trainset exist.

280 stars 24 forksSoftware Development Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add intertwine/dspy-agent-skills

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add intertwine/dspy-agent-skills --skill dspy-gepa-optimizer
Manual install
git clone --depth 1 https://github.com/intertwine/dspy-agent-skills.git /tmp/dspy-agent-skills
mkdir -p ~/.claude/skills
cp -r /tmp/dspy-agent-skills/skills/dspy-gepa-optimizer ~/.claude/skills/dspy-gepa-optimizer

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# DSPy GEPA Optimizer (3.2.x)

GEPA (Genetic-Pareto) is a reflective optimizer: it mutates a program's instructions and few-shots using an LM that reads your metric's **textual feedback** and proposes improvements. It maintains a Pareto frontier across validation tasks and is the default recommendation for complex DSPy workloads in 2026.

> The expansion "Genetic-Evolutionary Prompt Adaptation" that appears in some AI-generated summaries is an LLM-hallucinated backronym. The [paper](https://arxiv.org/abs/2507.19457) defines GEPA as Genetic-Pareto; the "Pareto" is load-bearing (GEPA keeps a frontier of candidates rather than collapsing to one).

## Prerequisites — do these first or GEPA wastes rollouts

1. A `dspy.Module` that runs end-to-end (see `dspy-fundamentals`).
2. A rich-feedback metric returning `dspy.Prediction(score=float, feedback=str)` (see `dspy-evaluation-harness`). Informative feedback can support reflection; evaluate optimizer benefit on the task rather than assuming superiority. A dict with the same fields still crashes `dspy.Evaluate` under DSPy 3.2.1 — use `dspy.Prediction`.
3. `trainset` and a **separate** `valset`. For GEPA, maximize training examples and keep validation just large enough to represent the downstream distribution; do not reuse the same examples for both.
Read the full skill on GitHub

Details

Skill name
dspy-gepa-optimizer
Path
skills/dspy-gepa-optimizer/SKILL.md
License
MIT
Repo updated
6 September 2026

FAQ

What does the Dspy Gepa Optimizer skill do?

Optimize DSPy programs with dspy.GEPA — a reflective/evolutionary optimizer to consider against task-specific baselines within an authorized evaluation budget. Use when the user says optimize, compile, GEPA, reflective optimization, or "make this program better" and a DSPy program + metric + trainset exist.

How do I install the Dspy Gepa Optimizer skill?

Claude Code plugin: /plugin marketplace add intertwine/dspy-agent-skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Dspy Gepa Optimizer popular?

It lives in intertwine/dspy-agent-skills, which has 280 GitHub stars and 24 forks.

Related skills

Dspy Advanced Workflow

280

intertwine/dspy-agent-skills

Build DSPy 3.2.x programs through spec, program, metric and baseline; extend to optimization and export when requested and justified by task budget. Orchestrates the other four DSPy skills (dspy-fundamentals, dspy-evaluation-harness, dspy-gepa-optimizer, dspy-rlm-module) in the correct order. Use for greenfield DSPy builds; prototypes may stop at a validated baseline.

Productivity & Notes24

Dspy Evaluation Harness

280

intertwine/dspy-agent-skills

Build DSPy evaluation harnesses with rich-feedback metrics that are essential for GEPA optimization. Use when writing a metric function, calling dspy.Evaluate, splitting dev/val sets, debugging "why is my optimizer not improving?", or designing CI-ready DSPy eval suites.

Writing & Content24

Dspy Fundamentals

280

intertwine/dspy-agent-skills

Write idiomatic DSPy 3.2.x programs — typed Signatures, dspy.Module subclasses, Predict/ChainOfThought/ReAct/ProgramOfThought, and save/load. Use this when starting any new DSPy project or when fixing non-idiomatic DSPy code (hard-coded prompts, ad-hoc string templates, untyped outputs, non-serializable classes).

Software Development24

Fix It Never Work Around It

365

NTCoding/claude-skillz

Stops execution and fixes root cause when commands, builds, scripts, or tools fail unexpectedly. Triggers on workaround language: 'directly', 'instead', 'alternatively', 'skip', 'fall back', 'work around', 'isn't working', 'broken', 'manually'. Activates on any unexpected non-zero exit code or process failure.

Software Development47

Lightweight Design Analysis

365

NTCoding/claude-skillz

This skill analyzes code for design quality improvements across 8 dimensions: Naming, Object Calisthenics, Coupling & Cohesion, Immutability, Domain Integrity, Type System, Simplicity, and Performance. Ensures rigorous, evidence-based analysis by: (1) Understanding code flow first via implementation-analysis protocol, (2) Systematically evaluating each dimension with specific criteria, (3) Providing actionable findings with file:line references. Triggers when users request: code analysis, design review, refactoring opportunities, code quality assessment, architecture evaluation.

Software Development47

Design Doc Mermaid

186

SpillwaveSolutions/design-doc-mermaid

Create Mermaid diagrams (flowchart, sequence, class, ER, state, C4, architecture) from text or source code. Default for GitHub wiki. Use when asked to create a diagram, generate mermaid, document architecture, or convert code to diagram. PlantUML is only for leftover types. Confluence needs PNG/SVG as well as the fence.

Software Development24

Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.