raveeshbhallaraveeshbhallaCommunityยท2 Agent Skills Included

dspy-gepa-logger

Observable prompt optimization with real-time dashboard monitoring

Optimizes language model prompts using the GEPA algorithm with full visibility into every iteration. Replaces blind prompt tuning with a live web dashboard showing prompt diffs, score lineage, and evaluation comparisons. Migrates existing DSPy GEPA code with a single import change and no other modifications.
npx skills add raveeshbhalla/dspy-gepa-logger --all -g -y

All Skills in This Repository (2)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install dspy-gepa-logger?โ–ผ

Run `npx skills add raveeshbhalla/dspy-gepa-logger --all -g -y` in your terminal to install all skills in this suite globally.

How to optimize prompts with GEPA?โ–ผ

Provide your dataset and a grading metric, then the skill builds a DSPy program and runs the GEPA optimizer to iteratively improve your prompt. You watch each candidate's score and feedback live in the web dashboard.

How to migrate from dspy.GEPA to gepa-observable?โ–ผ

Install the package with pip, then change your import from dspy.teleprompt to gepa_observable. All parameters, metrics, and compile calls work identically, and you can add dashboard monitoring with one server_url parameter.

Does it work with Claude Code and other coding agents?โ–ผ

Yes. The skills follow the universal SKILL.md standard and run in Claude Code, Codex, Cursor, and other compatible coding agents.

Can I use it without deep DSPy knowledge?โ–ผ

Yes. The gepa-demo skill guides you step by step, asking about your data fields, model choices, and budget, then generates and runs the optimization scripts for you.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’