WUPHF runs a collaborative office of AI agents on your machine. Set a goal, and they decompose tasks, hand off work, and surface blockers autonomously. It is free, open-source, and local-first.
Free
How to use WUPHF by Nex.ai?
Install with one command, open a browser at localhost:7891, and drop a goal in the #general channel. The CEO agent decomposes it, and ENG, DSG, CMO, and PM agents pick up tasks. They coordinate, resolve blockers, and ship work without human intervention. You can fork agent configs to customize.
WUPHF by Nex.ai 's Core Features
Autonomous coordination: Agents decompose goals, hand off tasks, and resolve dependencies without human routing. No prompt chaining needed.
Shared brain: All agents share a knowledge base that evolves as they work. Conclusions are promoted from individual notebooks to a team wiki.
Local-first: Runs entirely on your machine with no cloud dependency. Your data stays private and secure.
Free and open source: MIT licensed. No per-seat pricing, no usage fees, and no account required. Clone, build, and run.
Customizable agents: Each agent is a JSON config file. Fork the founding-team pack, swap tools, and integrate internal systems in hours.
Multi-provider support: Mix Claude Code, Codex, and OpenClaw agents in one office. They work together seamlessly.
Efficient: Uses 7x fewer tokens per session with fresh context windows, avoiding accumulated bloat.
WUPHF by Nex.ai 's Use Cases
Developers who want to ship features faster by delegating task decomposition and code reviews to AI agents.
Designers who need automated asset export and pixel-perfect coordination with engineering agents.
Product managers who can drop a goal and receive specs, PRs, and launch copy without manual follow-ups.
Startup founders who need a virtual team that works 24/7 without hiring or onboarding overhead.
Open source maintainers who want to automate issue triage, code contributions, and documentation updates.
