shiwenwen/hope-agent

🦭 A cross-device desktop AI agent with memory, autonomous goals, dynamic workflows, and headless deployment | 会记忆、能持续推进目标、会动态编排多 Agent 的跨端桌面 AI 助手，也可服务化常驻 NAS / 云端

Hope Agent – A Desktop‑first AI Assistant

What it is – Hope Agent is a Rust‑based, locally‑run AI assistant that lives on your computer (macOS, with experimental Windows/Linux builds) and can also be run head‑less as a service (Docker, NAS, cloud VM). It combines a native desktop GUI (built with Tauri 2 + React 19) with a web dashboard and optional IDE integration (ACP). The core idea is a personal agent that remembers you, pursues long‑term goals, and can act on the computer (run commands, edit files, control the browser) while keeping all data under your control.

Key capabilities

Area What Hope Agent does
Desktop experience One‑click installers, native macOS app, optional web UI at http://127.0.0.1:8420. Uses Tauri for a lightweight native wrapper around a React front‑end.
Goal‑driven work You define a Goal (desired outcome + criteria). The agent breaks it into a Workflow, runs it, and a Loop can automatically continue later. Supports pause/resume, budgeting, and audit of results.
Long‑term memory Persistent cross‑session memory stored locally (~/.hope-agent/). Supports fast recall, deep vector search, and periodic consolidation (“Dream Diary”).
Design & knowledge spaces Generates design artifacts (web pages, mobile mock‑ups, presentations, images, audio, etc.) with live preview, version history, and export to common formats. Provides a Markdown‑based “second brain” where you and the AI edit notes together, with backlinks and graph view.
Tool use & automation Can control macOS UI, browse the web, run shell commands, call built‑in AI image/audio generators, and invoke 40+ Feishu/Lark tools. All tool calls go through an approval flow; risky actions can be sandboxed in Docker.
Cross‑device hand‑off Same core can be accessed from desktop, a head‑less server, a web GUI, or IM channels (Telegram, Discord, Slack, Feishu, etc.). Sessions, memory, and task state travel with you.
Security & reliability Local‑first data, direct provider API calls, Docker sandbox for high‑risk tools, configuration snapshots with rollback, watchdogs for crash recovery, and optional bearer‑token auth for the server mode.
Extensibility Plugin‑style Hooks that trigger on 20+ lifecycle events, MCP client for transports/OAuth, and support for the agentskills.io skill format.

How to get started

  1. Install – macOS users can brew install --cask hope-agent or download the DMG; Windows via Scoop; Linux via AUR, apt, or dnf (experimental). Mirrors are provided at repo.hopeagent.ai for environments without GitHub access.
  2. Run – Launch the app (desktop GUI) or start the head‑less server with hope-agent server start. Open the web dashboard at http://127.0.0.1:8420.
  3. Create a Goal – Type a natural‑language outcome (e.g., “design a landing page for product Kanso”). The agent will generate a plan, run the necessary tools, and keep you updated.
  4. Connect tools – Grant macOS permissions for UI control, add your LLM API keys (or let Hope Agent install a local Ollama model), and optionally link an IM channel for remote hand‑off.

Who it’s for – Developers, designers, or power users who want an AI assistant that can do work (code, design, file manipulation) rather than just answer questions, while keeping data private and running on their own hardware.

Project health – CI runs on every push (GitHub Actions), releases are versioned with installers for multiple platforms, and a changelog is maintained. The repository is MIT‑licensed.

All details above are taken directly from the repository’s README.

Related

  • Project
    siddsachar/row-bot

    Row‑Bot is a self‑hosted desktop AI assistant that lets you converse with LLMs to create files, run code, manage designs, and interact with services—all while keeping data locally. It offers memory via a personal knowledge graph, goal‑driven agents, approval‑gated actions, multi‑device remote access, and integrations (GitHub, Notion, Slack, etc.). Runs on Windows/macOS/Linux or Docker, using Ollama or any OpenAI‑compatible provider, and stores everything securely on your machine.

  • Project
    fuyuxiang/echo-agent

    Echo Agent is an open‑source, self‑hosted AI‑agent runtime that provides long‑term memory, automatic skill extraction/evolution, and pluggable tool integration. It can run interactively in a terminal, as a long‑living Gateway service with a web dashboard and API, or be embedded in a desktop application (EchoAgent Desktop). The core runtime manages the agent loop, session context, model routing, permissions, and automation channels, while the memory system recalls past facts and the evolution system turns execution traces into reusable Skills. Install via `pip install "echo-agent[all]"`, configure an LLM API key, and start with `echo-agent run` or `echo-agent gateway`.

  • Project
    Pinvou/pinvou-agent

    Pinvou Agent is a cross‑platform desktop AI workspace built with Tauri 2, React and a Rust‑based agent engine (CodeWhale). It lets users attach files, run local or cloud LLMs, and use extensible tools to turn chat prompts into real deliverables—documents, designs, or code—while keeping data private when desired.

  • Project
    AgentiLoop/Agent

    AgentiLoop Agent is a native macOS app that lets any LLM act autonomously on your computer—coding, building, automating apps, running shell commands, fetching the web, sending iMessages, and responding to voice. It supports 23 providers, offers a self‑verifying task loop, sub‑agents, custom MCP tools, and a Swift‑based scripting system, all distributed as a signed, notarized binary.

  • Project
    zhayujie/CowAgent

    CowAgent is an open‑source, self‑hosted AI assistant that plans tasks, runs tools, keeps a personal knowledge base, and can operate as a team of cooperating agents. It supports many LLM providers, multimodal input, dozens of chat channels (Telegram, Slack, WeChat, etc.), a plug‑in skill system, and a three‑tier memory with nightly “Deep Dream” distillation. Install via a one‑line script or Docker, manage everything through a web console, and extend the system through the public Skill Hub or custom tool integrations.