jkf87jkf87Community·2 Agent Skills Included

ohmyclaw

Multi-model routing, orchestration, and account pooling for coding

Routes coding tasks across Z.ai GLM, OpenAI GPT, and OpenRouter models based on complexity and language. Runs Plan-Work-Review cycles with parallel workers, gap detection, and real-time channel notifications. Eliminates rate-limit stalls with multi-account pooling, cooldowns, and automatic fallback chains. Resumes interrupted long runs from saved checkpoints without repeating finished work.
npx skills add jkf87/ohmyclaw --all -g -y

All Skills in This Repository (2)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install ohmyclaw?

Run `npx skills add jkf87/ohmyclaw --all -g -y` in your terminal to install all skills in this repository globally.

What does ohmyclaw do?

It automatically picks the best model for each coding task and runs structured Plan-Work-Review cycles with code review and gap detection. It also tracks goal completion so long tasks finish reliably.

How does ohmyclaw avoid rate limits?

It spreads requests across multiple Z.ai, Codex, and Claude accounts using round-robin pooling with automatic cooldowns. If one account or provider fails, it falls back to the next healthy option.

Can ohmyclaw resume interrupted tasks?

Yes. Progress is committed to a local session store, so a crashed or timed-out run resumes with only the remaining work instead of starting over.

Does ohmyclaw support Korean language tasks?

Yes. It detects Korean-language requests and automatically prioritizes GLM models, which score highest for Korean coding and reasoning tasks.

Related Repositories in Software Engineering

View All in Software Engineering