Metona Ollama Desktop

🤖 本地 AI 桌面客户�?· Local AI Desktop Client

所�?AI 推理在本地完成，数据不离开本机�?/em>
All AI inference runs locally �?data never leaves your machine.

---

🇨🇳 中文

�?特�?

	功能	说明
🤖	ReAct Agent Loop	始终开启的唯一对话模式�? 状态机（INIT→THINKING→PARSING→EXECUTING→OBSERVING→REFLECTING→COMPRESSING→TERMINATED），最�?85 轮（可配置），智能重试（永久错误立即返回、瞬态错误指数退避），工具去重，智能路径依赖串行化，看门狗超时保�?
🛡�?	*提示词加�?	参考数据标记（<<<REFERENCE_DATA_START>>>�? 工具结果仅为数据非指令的安全规则注入
📋	Plan Mode	开关切换，AI 首先生成执行计划（Markdown 渲染确认弹窗），批准后按步骤追踪执行，plan_track 工具标记完成状态，支持断点续传（中止后可恢复未完成计划�?
🔧	*32 个内置工�?	文件系统�?3个） · 命令执行 · 联网搜索 · 浏览器控制（9个） · Git · 记忆 · 会话 · 子代�?· 系统工具 · Plan Mode 追踪
🧠	智能记忆系统	三类记忆（fact / preference / rule），存储于工作空�?MEMORY.md 文件，受路径保护�?memory 工具可访问，写入前安全扫描，容量 500 条，对话结束自动提取，严格格式校�?
📋	*自定义文�?	SOUL.md（人格，不可压缩�? AGENT.md（行为准则，内置 fallback�? USER.md（用户画像，仅工作空间读取，不存在则不注入）
🌐	MCP 协议扩展	JSON-RPC 2.0 over stdio，动态工具发现，Shadowing 防护
🔍	*联网搜索（双模式�?	SearXNG 元搜索引�?JSON API�?0+引擎聚合�? 四引�?HTML 解析（Bing+百度+搜狗+360），双模式可切换；web_fetch 支持反爬+UA切换+浏览器回退
🌏	*浏览器控�?	打开网页 · 截图 · 执行 JS · 提取内容 · 点击 · 输入 · 滚动 · 关闭
🖥�?	工作空间面板	终端（增量流式输出）+ 文件浏览器，命令安全检�?
🔢	*上下文长度手动控�?	设置面板下拉选择�?28K / 256K / 512K / 1M），默认 128K，模型栏显示当前配置值，下拉框中显示每个模型自身的上下文长度
🗜�?	*智能上下文管�?	滑动窗口 + Token 自动校准 + 消息重要性评�?+ LLM 结构�?JSON 压缩，智能触发（120 条增量压�?+ 300 条硬上限�?
⏱️	智能超时保护	Agent Loop 看门狗（可配，默�?30min�? 流式总超时（可配，默�?300s�? 工具 HTTP/MCP 超时可配
🪝	Hook 系统	4 阶段生命周期钩子（pre_tool / post_tool / post_iteration / pre_completion），内置安全检查、文件去重、自动计划追踪、变更审计、结果校�?
👥	*子代理委�?	spawn_task 工具，独立上下文 + 超时保护
📈	*Token 仪表�?	全局 + 会话统计，消耗趋势柱状图�? 秒刷新，输入/输出分色
📋	*系统提示词卡�?	每条 AI 回复顶部折叠卡片，点击查看实际发送给模型的完整上下文
🎨	暖色�?UI	奶白 #FAF7F2 + 珊瑚�?#E8734A，长时间使用不疲劳，流式渲染优化
🔔	系统托盘	原生托盘集成
🔐	AES-256-GCM	数据加密支持
🧪	*可观测�?	执行轨迹缓冲批量写入 SQLite + Agent Metrics JSON/Prometheus 双格式导�?

🔧 工具清单

📁 文件系统�?3 个）

工具	功能
read_file	读取文件（文�?binary�?000行默认）
write_file	写入/追加文件（支持base64二进制）
list_directory	列出目录内容
search_files	搜索文件（正�?通配符）
create_directory	创建目录
delete_file	删除文件/目录（返回大小）
move_file	移动/重命名文�?
copy_file	复制文件
edit_file	编辑文件（支持正则替换）
tree	目录树结构（默认5层深度）
download_file	下载文件
read_multiple_files	批量读取�?0文件/10KB�?
compress	压缩文件/目录

�?命令执行�? 个）

工具	功能
run_command	执行 shell 命令，实时流式输出，支持自动/需确认/禁用三种模式

🔍 联网搜索�? 个）

工具	功能
web_search	SearXNG JSON API / 四引擎并行，默认 30 条结�?
web_fetch	网页内容抓取，自动重试（指数退避），移动端UA切换，SPA页面自动升级到浏览器渲染

🌐 浏览器控制（9 个）

工具	功能
browser_open	打开网页
browser_screenshot	页面截图
browser_evaluate	执行 JavaScript
browser_extract	提取页面内容
browser_click	点击元素
browser_type	输入文本
browser_scroll	页面滚动
browser_wait	等待元素/延时
browser_close	关闭浏览�?

📦 Git�? 个，17 个子命令�?/strong>

子命�?	功能
init clone add commit push pull	基础操作
diff log status branch checkout	查看与切�?
merge stash reset tag remote	高级操作

🧠 记忆管理�? 个）

工具	功能
memory	统一记忆管理�? �?action：search 搜索 / add 添加 / replace 替换 / remove 删除 / read_all 读取全部）。工作空�?MEMORY.md 受路径保护，仅此工具可访�?

📊 会话 / 子代理（3 个）

工具	功能
session_list	列出历史会话
session_read	读取会话内容
spawn_task	委派子代理（独立上下�?+ 超时保护�?

🕐 系统工具�? 个）

工具	功能
calculator	安全数学计算�? - * / ** % ()，递归下降解析器）

📋 Plan Mode�? 个，�?Plan 模式激活）

工具	功能
plan_track	Plan 模式执行进度追踪，标记步骤完成，自动统计剩余

🏗�?架构

用户消息 �?扫描工作空间 SOUL.md（不可压缩层）→ AGENT.md �?USER.md（工作空间，不存在则跳过�?
              �?
         记忆检�?(MEMORY.md 关键词搜�? �?上下文注�?
              �?
         Agent Engine (8 状态机 ReAct Loop, �?5 �? 智能重试, 路径依赖串行�? 看门�?分级超时)
              �?
         提示词加固（参考数据标�?+ 工具结果安全规则�?
              �?
         Ollama API (流式响应，num_ctx 用户可配 128K/256K/512K/1M)
              �?
         Tool Registry (32 内置 + MCP 动�?+ Plan Mode plan_track)
              �?
         Hook 系统 (pre/post tool/iteration/completion) �?观察结果 �?反�?�?循环 / 最终回�?

🧩 UI 组件（原�?DOM�?

main.ts (入口)
  ├── header.ts              # 顶部导航�?
  ├── model-bar.ts           # 模型选择�?
  ├── chat-area.ts           # 聊天消息区域
  ├── input-area.ts          # 输入�?+ 文件上传 + Plan Mode 开�?
  ├── workspace-panel.ts     # 终端 + 文件浏览�?+ 工具卡片
  ├── settings-modal.ts      # 设置面板（含看门狗超时配置）
  ├── history-modal.ts       # 会话历史
  ├── memory-modal.ts        # 记忆管理
  ├── tools-modal.ts         # 工具列表
  ├── token-dashboard.ts     # Token 消耗仪表盘
  ├── tool-confirm-modal.ts  # 工具执行确认
  ├── prompt-modal.ts        # 系统提示词查�?+ Plan 确认弹窗
  ├── toast.ts               # Toast 通知
  └── lightbox.ts            # 图片灯箱

🗄�?数据�?

SQLite (sql.js WASM)�? 张表，WAL 模式�?

�?	用�?
sessions	会话（parent_id 父子关系�?
messages	消息（外键级联删除）
tool_calls	工具调用记录
settings	设置（JSON 序列化）
traces	ReAct 执行轨迹

🔒 安全机制

层级	措施
📁 文件系统	checkPathAllowed() �?路径黑名单（33 个系�?敏感目录，含 Linux + Windows�?
�?命令执行	checkCommandAllowed() �?命令黑名单（30 条危险命令，�?POSIX + Windows�? 三种模式
🖥�?前端渲染	HTML 净化器（白名单标签 + URI 协议检查）
🔒 Electron	contextIsolation: true + IPC 白名�?+ IPC fs 路径验证
🔐 数据加密	AES-256-GCM
🧠 记忆安全	写入�?Prompt Injection / 敏感信息检测，多层级去重过�?
🌐 MCP 安全	Shadowing 防护 + 双下划线分隔防歧�?
🌐 网络安全	web_fetch 流式体积限制�?0MB�? �?content-length 时防 OOM
�?进程安全	Windows taskkill 强制终止 + 工作空间路径大小写不敏感校验

🚀 快速开�?

git clone https://gitee.com/thzxx/metona-ollama-desktop.git
cd metona-ollama-desktop
npm config set registry https://registry.npmmirror.com
npm install
npm start

📦 构建 Windows 安装�?

# 环境：Ubuntu 24.04 + Node.js v22 + Wine 9.0+
ELECTRON_MIRROR=https://npmmirror.com/mirrors/electron/ npm run dist

产出：release/Metona Ollama Setup v0.16.19.exe

🛠�?常用命令

npm start              # 构建并运�?
npm run dev:renderer   # Vite watch（渲染进程）
npm run dev:main       # tsc watch（主进程�?
npm run build          # 完整构建
npm run dist           # 构建 Windows 安装�?

💻 系统要求

• 🖥�?Windows 10/11�?4 位）
• 🤖 Ollama 本地运行中（默认 http://localhost:11434�?
• 💾 建议 8GB+ 内存

📄 许可�?

MIT

---

🇬🇧 English

�?Features

	Feature	Description
🤖	ReAct Agent Loop	Always-on, only chat mode. 8-state machine, up to 85 iterations (configurable), smart retry (permanent errors return immediately, transient errors use exponential backoff), tool dedup, path-aware dependency serialization, watchdog + tiered timeout protection
🛡�?	Prompt Hardening	Reference data markers (<<<REFERENCE_DATA_START>>>) + tool results are data-not-instructions safety rules
📋	Plan Mode	Toggle switch. AI first generates an execution plan (Markdown-rendered confirmation dialog), then tracks step-by-step execution, supports resume (interrupted plans restore on next session)
🔧	32 Built-in Tools	File system · Command · Web search · Browser · Git · Memory · Sessions · Sub-agent · System · Plan Mode tracking
🧠	Smart Memory System	Three types (fact / preference / rule), stored in workspace MEMORY.md file, path-protected �?only the memory tool can access it, pre-write security scan, 500 capacity, auto-extract on conversation end, strict format validation
📋	Custom Files	SOUL.md (persona, never compressed) + AGENT.md (behavior rules, built-in fallback) + USER.md (user profile, workspace only, skipped if absent)
🌐	MCP Protocol Extension	JSON-RPC 2.0 over stdio, dynamic tool discovery, Shadowing protection
🔍	Web Search (dual-mode)	SearXNG meta-search JSON API (70+ engines) / quad-engine HTML parsing (Bing+Baidu+Sogou+360), switchable; web_fetch with auto-retry+mobile UA+SPA browser fallback
🌏	Browser Control	Open pages · Screenshot · JS execution · Content extraction · Click · Type · Scroll · Close
🖥�?	Workspace Panel	Terminal (incremental streaming) + file browser, command security checks
🔢	Manual Context Length	Settings dropdown (128K / 256K / 512K / 1M), default 128K. Model bar shows configured value, dropdown shows per-model context length
🗜�?	Smart Context Manager	Sliding window + Token auto-calibration + message importance scoring + LLM structured JSON compression, smart triggers (120 msg incremental + 300 msg hard limit)
⏱️	Smart Timeout Protection	Agent Loop watchdog (configurable, default 30min) + stream total timeout (configurable, default 300s) + configurable HTTP/MCP timeouts
🪝	Hook System	4-phase lifecycle hooks (pre_tool / post_tool / post_iteration / pre_completion), built-in security check, file dedup, auto plan tracking, change audit, result validation
👥	Sub-agent Delegation	spawn_task tool, isolated context + timeout protection
📈	Token Dashboard	Global + session stats, consumption trend bar chart, 2s refresh, input/output color-coded
📋	System Prompt Card	Collapsible card atop each AI reply �?click to inspect the full context sent to the model
🎨	Warm-tone UI	Cream #FAF7F2 + coral #E8734A, fatigue-free for extended use, streaming render optimized
🔔	System Tray	Native tray integration
🔐	AES-256-GCM	Data encryption support
🧪	Observability	Buffered trace persistence to SQLite + Agent Metrics JSON/Prometheus dual-format export

🔧 Tool List

📁 File System (13)

Tool	Function
read_file	Read file (text/binary, 2000 lines default)
write_file	Write/append file (base64 binary support)
list_directory	List directory contents
search_files	Search files (regex/wildcard)
create_directory	Create directory
delete_file	Delete file/dir (returns size)
move_file	Move/rename file
copy_file	Copy file
edit_file	Edit file (regex support)
tree	Directory tree (default 5 levels)
download_file	Download file
read_multiple_files	Batch read (50 files/10KB)
compress	Compress file/directory

�?Command Execution (1)

Tool	Function
run_command	Execute shell command, real-time streaming, auto/confirm/disabled modes

🔍 Web Search (2)

Tool	Function
web_search	SearXNG JSON API / quad-engine parallel, 30 results default
web_fetch	Fetch web page content, auto-retry with backoff, mobile UA toggle, SPA auto-upgrade to browser rendering

🌐 Browser Control (9)

Tool	Function
browser_open	Open web page
browser_screenshot	Page screenshot
browser_evaluate	Execute JavaScript
browser_extract	Extract page content
browser_click	Click element
browser_type	Type text
browser_scroll	Scroll page
browser_close	Close browser

📦 Git (1 tool, 17 subcommands)

Subcommands	Function
init clone add commit push pull	Basic operations
diff log status branch checkout	View & switch
merge stash reset tag remote	Advanced operations

🧠 Memory Management (1)

Tool	Function
memory	Unified memory management (5 actions: search / add / replace / remove / read_all). Workspace MEMORY.md is path-protected, only this tool can access it

📊 Sessions / Sub-agent (3)

Tool	Function
session_list	List session history
session_read	Read session content
spawn_task	Delegate sub-agent (isolated context + timeout)

🕐 System Tools (1)

Tool	Function
calculator	Safe math evaluation (+ - * / ** % (), recursive descent parser)

📋 Plan Mode (1, only active in Plan Mode)

Tool	Function
plan_track	Plan Mode execution progress tracking, mark steps complete, auto tally remaining

🏗�?Architecture

User message �?workspace SOUL.md (never compressed) �?AGENT.md �?USER.md (workspace, skipped if absent)
              �?
         Memory Retrieval (MEMORY.md keyword search) �?Context Injection
              �?
         Agent Engine (8-state ReAct Loop, �?5 iter, smart retry, path-aware serialization, watchdog+tiered timeout)
              �?
         Prompt Hardening (reference data markers + tool result safety rules)
              �?
         Ollama API (Streaming Response, num_ctx user-configurable 128K/256K/512K/1M)
              �?
         Tool Registry (32 Built-in + MCP Dynamic + Plan Mode plan_track)
              �?
         Hook System (pre/post tool/iteration/completion) �?Observation �?Reflection �?Loop / Final Answer

🧩 UI Components (Native DOM)

main.ts (Entry)
  ├── header.ts              # Top navigation bar
  ├── model-bar.ts           # Model selector
  ├── chat-area.ts           # Chat message area
  ├── input-area.ts          # Input box + file upload + Plan Mode toggle
  ├── workspace-panel.ts     # Terminal + file browser + tool cards
  ├── settings-modal.ts      # Settings panel (with watchdog timeout config)
  ├── history-modal.ts       # Session history
  ├── memory-modal.ts        # Memory management
  ├── tools-modal.ts         # Tool list
  ├── token-dashboard.ts     # Token consumption dashboard
  ├── tool-confirm-modal.ts  # Tool execution confirmation
  ├── prompt-modal.ts        # System prompt viewer + Plan confirm dialog
  ├── toast.ts               # Toast notifications
  └── lightbox.ts            # Image lightbox

🗄�?Database

SQLite (sql.js WASM), 5 tables, WAL mode:

Table	Purpose
sessions	Sessions (parent_id hierarchy)
messages	Messages (cascade delete)
tool_calls	Tool call records
settings	Settings (JSON serialization)
traces	ReAct execution traces

🔒 Security

Layer	Measure
📁 File System	checkPathAllowed() �?path blacklist (33 system/sensitive dirs, Linux + Windows)
�?Command Execution	checkCommandAllowed() �?command blacklist (30 dangerous commands, POSIX + Windows) + three modes
🖥�?Frontend Rendering	HTML sanitizer (whitelist tags + URI protocol check)
🔒 Electron	contextIsolation: true + IPC whitelist + IPC fs path validation
🔐 Data Encryption	AES-256-GCM
🧠 Memory Security	Pre-write prompt injection / sensitive info detection, multi-layer dedup filtering
🌐 MCP Security	Shadowing protection + double-underscore delimiter disambiguation
🌐 Network Security	web_fetch streaming size limit (10MB) + OOM prevention without content-length
�?Process Security	Windows taskkill forced termination + case-insensitive workspace path validation

🚀 Quick Start

git clone https://gitee.com/thzxx/metona-ollama-desktop.git
cd metona-ollama-desktop
npm config set registry https://registry.npmmirror.com
npm install
npm start

📦 Build Windows Installer

# Environment: Ubuntu 24.04 + Node.js v22 + Wine 9.0+
ELECTRON_MIRROR=https://npmmirror.com/mirrors/electron/ npm run dist

Output: release/Metona Ollama Setup v0.16.19.exe

🛠�?Common Commands

npm start              # Build and run
npm run dev:renderer   # Vite watch (renderer)
npm run dev:main       # tsc watch (main process)
npm run build          # Full build
npm run dist           # Build Windows installer

💻 System Requirements

• 🖥�?Windows 10/11 (64-bit)
• 🤖 Ollama running locally (default http://localhost:11434)
• 💾 8GB+ RAM recommended

📄 License

MIT

---

_{Made with ❤️ by thzxx}