xuyxuyCommunity·1 Agent Skills Included

sgo

Test products, resumes, and pitches against synthetic audiences

Evaluates products, resumes, pitches, and policies against realistic census-grounded persona panels. Replaces slow surveys and guesswork with fast audience scoring, segmented feedback, and ranked improvement suggestions. Runs counterfactual probes to show exactly which changes would win over undecided evaluators, plus bias audits to keep results realistic.
npx skills add xuy/sgo --all -g -y
Available:

Guides the agent through the full six-phase pipeline, from setup and entity definition to cohort building, evaluation, counterfactual probing, iteration, and bias auditing, with exact questions to ask the user at each decision point.

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install SGO?

Run `npx skills add xuy/sgo --all -g -y` in your terminal to install it globally. You will also need an OpenAI-compatible API key in a .env file.

What is Semantic Gradient Optimization?

It is a method that scores your product, resume, or pitch against a panel of realistic synthetic personas, then tests hypothetical changes to rank which ones would improve reception the most.

How does SGO replace traditional surveys?

Instead of recruiting real people over weeks, SGO builds a stratified panel from a 1M-person census-grounded dataset and gets scored feedback in minutes for pennies.

Can I test my resume or investor pitch with SGO?

Yes. SGO works for anything people evaluate, including resumes, pitch decks, landing pages, policies, and content, with audience segments tailored to each domain.

Does SGO work with Claude Code?

Yes. It follows the standard SKILL.md format and runs as an interactive skill in Claude Code and other compatible coding environments.

Related Repositories in Marketing & Sales

View All in Marketing & Sales