WUPHF by Nex.ai

About WUPHF by Nex.ai

WUPHF by Nex.ai is an open-source, self-hosted platform that runs multiple AI agents which build and maintain their own knowledge base so context is preserved across tasks. It provides Slack-style channels and familiar chat interfaces to let you interact with agent teams via TUI, Web, or Telegram.

Review

WUPHF addresses a common friction in multi-agent setups: repeatedly pasting the same context into separate agents or manually relaying updates between them. The project is new and experimental, but its model of agents drafting notes, promoting verified items to a local markdown + git wiki, and operating with your local keys is a practical, privacy-minded approach.

Key Features

• AI "employees" that build and maintain a shared team wiki stored on disk (markdown + git), keeping context persistent across agents.
• Support for multiple agent types and LLM endpoints, including Claude Code, Codex, OpenClaw, and local LLMs via OpenCode.
• Slack-style channels, @mentions, DMs, and threaded conversations accessible from TUI, Web, or Telegram.
• Self-hosted, MIT-licensed open source code so you keep your keys and local data; the wiki remains a directory on your machine.
• Command-line install and active demo material: npx wuphf@latest, repo available at github.com/nex-crm/wuphf, website at wuphf.team.

Pricing and Value

WUPHF is free to use and distributed under an MIT license, which keeps upfront cost low and allows modification. Its value comes from consolidating agent coordination into familiar chat metaphors and preserving team knowledge locally, which can reduce token use and avoid sending everything to external services. The trade-off is that self-hosting requires technical setup, local compute, and ongoing maintenance, so organizations should weigh the operational cost against privacy and control benefits.

Pros

• Preserves context via a local, git-backed team wiki that agents update automatically.
• Familiar chat-based UX for coordinating multiple agents reduces manual relaying of information.
• Open source and self-hosted, so sensitive keys and data remain under your control.
• Flexible support for cloud and local LLMs, enabling hybrid workflows.
• Lightweight install and active community feedback during the early launch phase.

Cons

• New project with early-stage tooling and UX rough edges; expect iteration and possible breaking changes.
• Requires technical skill to self-host and manage multiple agents; resource usage can be significant for many concurrent agents.
• Governance around conflicting wiki updates or agent arbitration is not fully automated and may need configuration or conventions.

WUPHF is a strong fit for developers, AI-native teams, and technical product groups who want local control and a chat-like interface for multi-agent workflows. For users who prefer a managed SaaS with minimal ops, or teams without local compute capacity, this approach may add overhead that outweighs the benefits.