✍️

Writing Style AutoResearch Agent

contentapi-pushhours cycle

Autonomously optimize a writing style guide by iterating variants against reference samples using LLM-as-judge scoring overnight.

Primary Metric

Style Match Score

↑ Higher is better

Experiment Settings

Variants per round: 5

Max rounds: 15

Min sample size: 200

Guard Rails
Experiments auto-pause if these thresholds are breached
Minimum 30% style match scoreFloor: 30.0%
Start Experiment →