---
title: "gepa-omni"
description: "Optimize scorable prompts, code, configs, schemas, SQL, regexes, and agent instructions with GEPA's Omni portfolio and fresh continuation workflow."
canonical: https://agentpluginsdirectory.com/plugins/gepa-omni
last-updated: 2026-09-22
---

# gepa-omni
Optimize scorable prompts, code, configs, schemas, SQL, regexes, and agent instructions with GEPA's Omni portfolio and fresh continuation workflow.
- Slug: gepa-omni
- Publisher: Qredence
- Repository: https://github.com/Qredence/gepa-omni
- Manifest: plugin.json
- Version: 0.2.1
- License: MIT
- Category (editorial): agent-tooling
- Skills: 1 (gepa-omni-skill)
- MCP servers: 0
- Stars: 3
- Repository created: 2026-08-04
- Repository last pushed: 2026-09-09
- Publisher type: Organization
- Listing: https://agentpluginsdirectory.com/plugins/gepa-omni
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What gepa-omni does, in the publisher's words

Optimize any scorable text artifact, prompts, programs, configurations, schemas, SQL, regular expressions, plans, or agent instructions, from plain evaluator feedback.

GEPA Omni packages the GEPA Anything optimization stack as an Agent Plugins 1.0 plugin, so any compatible coding agent can install it as a skill and drive it for you. It ships the standalone reflective GEPA engine from PyPI plus three plugin-native engines, AutoResearch, Meta-Harness, and Best-of-N, and runs them together in a two-phase Omni workflow.

- One evaluator in, a better artifact out. Write a function that scores a candidate and explains failures; the engines handle mutation, selection, and budgeting.
- Four engines, one contract. GEPA, AutoResearch, Meta-Harness, and Best-of-N all run against the same candidate, data, and budget.
- Omni by default. Phase 1 explores with three engines in parallel; Phase 2 continues the best candidate with a fresh optimizer.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/Qredence/gepa-omni/HEAD/README.md

## Skills

- gepa-omni-skill: Automatically improve any scorable text artifact, prompts, programs, configs, specs, regex/SQL/schemas, agent scaffolds, or encoded search solutions, with the engine-pluggable optimize_anything API. Use when tuning a candidate from objective metrics, execution feedback, or LLM-as-judge scores, when d…

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.