Hi, I'm Zehua Hao.
I build around AI agents, coding workflows, CLI-first model integration, and developer tooling.
- PhD student in Computer Science
- Research and engineering interests:
- AI agents and multi-agent systems
- coding assistants and workflow automation
- retrieval, reasoning, and tool-augmented systems
- turning CLI-first model access into reusable engineering workflows
- Based in Xi'an, China
- LikeCode / Claude Code restoration and analysis
- reverse engineering, source analysis, runtime verification, and workflow reconstruction
- Everything Agent-CLI to Claude Code
- bringing Gemini, Qwen, Codex, Grok, and other CLI-first capabilities back into a Claude Code-centered workflow
- Multi-model review workflows
- using different models for planning, critique, implementation review, and document analysis
- learn-likecc
- Claude Code reverse engineering, source-map analysis, build recovery, and workflow notes
- everything-agent-cli-to-claude-code
- umbrella repo for
usecli:*workflows, plugin specs, and provider integration experiments
- umbrella repo for
- gemini-plugin-cc
- Gemini-oriented plugin scaffold for the Claude Code ecosystem
- qwen-plugin-cc
- Qwen-oriented plugin scaffold with auth-aware workflow direction
- grok-plugin-cc
- Grok-oriented plugin scaffold for critique and review workflows
- trae-plugin-cc
- Trae-oriented plugin scaffold for future integration work
- making multi-model engineering workflows practical
- converting external AI CLIs into reusable skills and wrappers
- building cleaner bridges between research workflows and production developer tools
I'm especially interested in collaborating on:
- agent engineering
- coding assistant workflows
- model orchestration
- CLI integration and automation
- research tooling for paper review and technical analysis
- GitHub: @Harzva
- ORCID: https://orcid.org/0009-0006-8057-2958
This profile README is intentionally focused on the projects and workflows I'm actively building right now.