Chat Assistants

cue is an open-source, bring-your-own-key macOS copilot that watches your screen, listens to your meetings, and tries to stay hidden from screen-capture tools.

Hexabot v3 is a TypeScript runtime that keeps agentic workflows on rails by binding them to YAML definitions, Zod schemas, and explicit memory across chat channels.

An open-source AI assistant built in Kotlin that remembers context across chats, draws interactive screens, and runs a sandboxed Linux distro on your phone.

Open-source client code for hardware makers who want real-time voice AI without building the backend themselves.

OpenMinis exists to turn your smartphone into a private AI workstation with a real Linux shell, system integrations, and your own API keys.

It fixes web chatbots' amnesia and censorship by bundling local LLMs, voice chat, and Live2D avatars into one offline Windows app.

Rocketnotes pairs a hierarchical note editor with an AI agent that reads your scattered snippets and archives them where they belong.

Nativ bundles an embedded mlx-vlm server into a SwiftUI app to turn your Apple Silicon Mac into a local AI workspace with OpenAI-compatible APIs.

DEEIX Chat exists because most open-source AI frontends stop at the fun stuff and leave you to invent user management, billing, and audit logs.

TauriTavern ports the SillyTavern LLM frontend into a native Tauri v2 desktop app, swapping browser HTTP calls for Rust-backed commands while keeping the upstream UI intact.

Powers an open workplace by treating AI agents as native conversation participants with built-in orchestration, not bolted-on bots.

PenEcho exists because translating a half-formed diagram into a chat prompt is a waste of keystrokes.

Aether wraps local LLM tool-calling in a polished Android, iOS, and macOS interface, aiming to outshine the usual open-source mobile aesthetic.

It exists to hide a real-time Gemini copilot from Zoom, Meet, and OBS while you work through technical interviews.

RAG Web UI bundles document ingestion, vector search, and multi-provider LLM dialogue into a single dockerized stack so you don't have to wire FastAPI and LangChain together for the hundredth time.

Hermex is a free, native SwiftUI client that turns your iPhone into a control panel for a self-hosted Hermes agent, keeping the compute on your hardware and the controls in your pocket.

Hope Agent wants to be the desktop AI that actually remembers what you were doing when you switch from Mac to phone to browser.

ChatbotX is a self-hosted ManyChat alternative that wires AI agents into omnichannel chat flows so marketing teams keep their contact lists off someone else’s servers.

Proma wraps the Claude Agent SDK into an Electron workbench that keeps your data in local JSONL files and lets colleagues trigger file-editing agents from Feishu group chats.

Built for sellers tired of manually haggling, shipping, and juggling multiple accounts on China’s largest second-hand marketplace.