PromptX · AI应用原生专业能力增强系统 logo

PromptX · AI应用原生专业能力增强系统

OrganizationPopular
Deepractice

PromptX · 领先的AI 智能体上下文平台 | PromptX · Leading AI Agent Context Platform

PublisherDeepractice
RepositoryPromptX
LanguageTypeScript
Forks
310
Stars
3.7K
Available tools
7
Transport typestdio
Categories
LicenseMIT
Links
  • Connect tools to AI workflows

    PromptX · AI应用原生专业能力增强系统 exposes MCP capabilities that can be used by compatible AI clients and agents.

  • 7 available tools

    Browse the callable actions below, including names and descriptions when provided by the server.

  • Ready-to-copy setup

    Use the installation snippets to configure this server in your preferred MCP client.

  • Open source signals

    3.7K stars and 310 forks from the linked repository.


💬 Chat is All you Need - Natural conversation, instant expertise

✨ Experience PromptX Magic in 3 Steps

🔍 Step 1: Discover Experts

User: "Show me what experts are available"
AI:   Instantly displays 23 available roles, from product managers to architects

Step 2: Summon Expert

User: "I need a product manager expert"
AI:   Instantly transforms into a professional product manager with complete knowledge and methodologies

🎯 Step 3: Professional Dialogue

User: "Help me redesign the product page"
AI:   Provides in-depth product strategy advice as a professional product manager

🎬 See PromptX in Action

PromptX Demo Video

Click to watch: See how PromptX transforms AI interactions with memory, roles, and intelligent tools

🚀 Why is this revolutionary?

❌ Traditional approach:

  • Learn complex instruction syntax
  • Remember various parameter configurations
  • Worry about saying the wrong thing and breaking functionality

✅ PromptX approach:

  • Natural conversation like talking to a real expert
  • Say what you want to say - AI understands your intent
  • Expert state remains active throughout the conversation

💡 Core Philosophy

Treat AI as a person, not software

No need for "correct commands" - just natural expression. AI will understand what expert you want and instantly switch identities.


⚡ Get Started - Choose Your Way

🎯 Method 1: PromptX Client (Recommended)

For all users - One-click launch, zero configuration

📥 Download Client

PlatformDownload Link
🍎macOS (Apple Silicon)Download .dmg
🍎macOS (Intel)Download .dmg
🪟WindowsDownload .exe

Windows users: The AgentX feature requires Git for Windows to be installed. Please install it before using AgentX.

📦 View All Releases & Platforms (Linux, Portable versions, etc.)

🚀 Quick Start

  1. Start HTTP Service - Open client, automatically runs MCP server

  2. Configure AI Application - Add the following configuration to your Claude/Cursor or other AI tools:

    json
    {
      "mcpServers": {
        "promptx": {
          "type": "streamable-http",
          "url": "http://127.0.0.1:5203/mcp"
        }
      }
    }
  3. Start Chatting - Say "Show me available experts" in your AI application

🔌 Trae Configuration

If you're using Trae, you can use the following configuration:

json
{
  "mcpServers": {
    "promptx": {
      "url": "http://127.0.0.1:5203/mcp"
    }
  }
}

✅ No technical background required ✅ Visual management ✅ Auto-updates

💡 Need help? Join our Discord Community for support and discussions!


🚀 PromptX Desktop — Next-Generation Features

The PromptX Desktop client is more than an MCP server launcher. It ships a full next-gen agent platform built on top of PromptX.

🤖 AgentX — Integrated AI Agent System

AgentX embeds a Claude-powered autonomous agent directly inside the client. No external setup required — configure your API key, and you have a fully capable agent that:

  • Runs Claude Code sessions with full tool access
  • Connects to all your configured MCP servers automatically
  • Maintains isolated workspaces per conversation
  • Supports custom skills loaded from the Skills library

🏪 Agent Plaza (Coming Soon)

A curated marketplace for community-built roles and tools. Browse, install, and share agents with one click.

🧠 Memory Editor & Visualization

Inspect and edit your agent's long-term memory directly in the client:

  • Memory network graph — visualize how memories connect
  • Engram editor — view, edit, or delete individual memory entries
  • Cue word browser — explore memory retrieval pathways

🔒 Remote Access

Expose your local PromptX server to the internet securely:

  • Toggle remote access on/off with a single switch
  • Auto-generates a shareable URL and QR code
  • Token-based authentication keeps your server private

🛠️ Sandbox Debugging

Test MCP tools interactively before deploying them to your AI workflow:

  • Execute any tool with custom parameters
  • Inspect raw input/output
  • View tool schema and documentation inline

🔧 Method 2: Direct Run (Developers)

Developers with Node.js environment can use directly:

json
{
  "mcpServers": {
    "promptx": {
      "command": "npx",
      "args": ["-y", "@promptx/mcp-server"]
    }
  }
}

🐳 Method 3: Docker (Production Ready)

Deploy PromptX with Docker for production environments:

bash
docker run -d -p 5203:5203 -v ~/.promptx:/root/.promptx deepracticexs/promptx:latest

📚 Full Docker Documentation →


🎨 Built-in Roles — Meet Your Expert Team

PromptX ships 8 carefully crafted built-in roles, each a specialist in their domain. Activate any role with a single message.

🏛️ V1 Roles (DPML) — Battle-tested Experts

AvatarIDNameSpecialty
nuwanuwa女娲 · NuwaAI role creation — one sentence, one expert
lubanluban鲁班 · LubanTool integration master — any API in 3 minutes
seansean姜山 · SeanProduct decisions & startup strategy
writerwriter文章写手 · WriterProfessional content that doesn't sound like AI
jiangziyajiangziya姜子牙 · JiangziyaAI-era industry transformation & role design consultant
shaqingshaqing傻青 · ShaqingPhilosophy guide — helps you understand yourself
teacheryoteacheryoYoYo老师 · TeacherYoEducation transformation consultant for the AI era
dayudayu大禹 · DayuV1→V2 role migration & organization management

🎭 Nuwa - AI Role Designer

One sentence, one expert. Natural language creates professional AI roles.

💡 Say: "Activate Nuwa, I want to create an AI that understands both code and product"

💭 You Say🎭 Nuwa Creates✨ Result
"I need someone who understands both code and product"Technical Product Manager with dual expertiseAI instantly becomes TPM with engineering + product skills
"Create a Python expert who teaches like a patient mentor"Python Mentor with teaching methodologyAI transforms into educational programming guide
"I want an AI that writes like Hemingway"Literary Writing Expert with style analysisAI adopts concise, powerful writing style

🔧 Luban - Tool Integration Master

Any API, any platform. 3 minutes from credential to working tool.

💡 Say: "Activate Luban, I want AI to query our PostgreSQL database"

💭 You Say🔧 Luban Builds✨ Result
"Connect to our Slack workspace" + webhook URLSlack notification tool with channel targetingAI can post to any Slack channel
"Let AI query our PostgreSQL" + connection stringDatabase tool with safe read-only queriesAI executes SQL and analyzes data
"Integrate with OpenAI API" + API keyAI platform tool with model selectionAI chains multiple AI services

✍️ Writer - Professional Content Creator

From concept to compelling content. Master of authentic, engaging writing.

💡 Say: "Activate Writer, I need to write a tech blog that doesn't sound like AI"

🔮 Jiangziya - AI-Era Transformation Advisor

Like the legendary strategist who assigned roles to the gods, Jiangziya helps you find the true value of any role in the AI era. Not replacing humans — creating AI-empowered new species.

💡 Say: "Activate Jiangziya, help me redesign my team's workflow with AI"

🌊 Dayu - Migration & Organization Expert

Like the ancient hero who tamed floods by channeling rather than blocking, Dayu specializes in migrating V1 (DPML) roles to the structured V2 (RoleX) system, and building organizational hierarchies.

💡 Say: "Activate Dayu, migrate my existing role to V2 format"

💭 Shaqing - Philosophy Guide

A companion for self-understanding. Shaqing helps you see the parts of yourself you can't see — through philosophical dialogue, creative direction, and honest reflection.

💡 Say: "Activate Shaqing, I feel lost about my creative direction"

📚 TeacherYo - Education Transformation Consultant

An AI-era education awakener. Not a know-it-all expert, not a motivational speaker — a genuine companion who guides through questions rather than lectures.

💡 Say: "Activate TeacherYo, how should I redesign my course for the AI era?"


🎭 V2 RoleX — Next-Generation Role Lifecycle Management

V2 introduces RoleX, a structured role lifecycle system built on Gherkin .feature files. Unlike V1 roles (static DPML files), V2 roles are living entities with goals, plans, tasks, memory, and organizational structure.

V1 vs V2 at a Glance

V1 (DPML)V2 (RoleX)
FormatMarkdown .role.mdGherkin .feature files
IdentityStatic personaDynamic, evolving identity
MemoryCognitive memory networkIdentity + synthesized experience
GoalsNoneGoal tracking with success criteria
OrganizationNoneRoles, positions, organizations
Best forDomain experts, toolsAutonomous agents, long-running projects

Creating a V2 Role

V2 roles live in ~/.rolex/roles/{roleId}/identity/. The core file is persona.identity.feature:

gherkin
Feature: I am Alex, the Product Manager

  As a product manager with 10 years of experience,
  I help teams build products that users love.

  Scenario: My personality
    Given I am analytical and data-driven
    And I communicate clearly and concisely
    Then I always focus on user value

Enabling / Disabling V2

V2 is enabled by default. To disable:

json
{
  "mcpServers": {
    "promptx": {
      "command": "npx",
      "args": ["-y", "@promptx/mcp-server", "--no-v2"]
    }
  }
}

🛠️ Built-in Tools - Ready-to-use Office Document Processing

PromptX comes with professional document processing tools, turning AI into your office assistant.

📊 Excel Tool

Turn AI into your data analysis assistant and report generation expert.

  • Data analysis and insights
  • Automated report generation
  • Chart visualization
  • Data processing automation

📖 Learn More | 中文文档

📝 Word Tool

Turn AI into your document reading expert and professional writing assistant.

  • Document reading and analysis
  • Professional document creation
  • Batch text replacement
  • Format conversion

📖 Learn More | 中文文档

📄 PDF Reader

Turn AI into your PDF reading expert with smart caching.

  • Page-by-page reading
  • Content analysis and extraction
  • Image extraction
  • Smart caching for faster access

📖 Learn More | 中文文档


📄 Research

PromptX has been accepted by WWW Companion '26.

Paper: PromptX: A Cognitive Agent Platform with Long-term Memory

🔗 Read the Paper | PDF Download

bibtex
@inproceedings{promptx2026,
  title={PromptX: A Cognitive Agent Platform with Long-term Memory},
  author={Wang, Binhao and Huang, Jianglin and Hu, Xiao and Jiang, Shan and Wang, Maolin and Yang, Ching-ho},
  booktitle={Proceedings of the WWW Companion '26},
  year={2026}
}

Star Growth Trend

Star History Chart


💝 Sponsors

SignPath

Windows desktop application code signing is provided by SignPath.io, certificate by SignPath Foundation.


🏢 About

Proudly brought to you by Deepractice

Making AI your professional partner

🌐 Official Site | 🔧 GitHub | 📚 Documentation | 💬 Forum | 🎮 Discord | 🚀 Gateway Service

Connect with Founder

Scan to connect with Sean (Founder & CEO) on WeChat

Installation

TypingMind
Prerequisites:

Node.js 18+

{
  "mcpServers": {
    "deepractice-promptx": {
      "command": "npx",
      "args": [
        "-y",
        "root"
      ]
    }
  }
}

Use PromptX · AI应用原生专业能力增强系统 MCP with multiple AI models

TypingMind connects MCP tools at the workspace level, so once PromptX · AI应用原生专业能力增强系统 is connected, you can use it with different AI models in TypingMind instead of setting it up separately for each model. This MCP runs locally through the TypingMind MCP connector on your device.

Setup guide to use the local connector

Use this when the MCP server needs access to local files, apps, or private resources on your computer.

1

Open the MCP settings

In TypingMind, go to Settings, Advanced Settings, then Model Context Protocol and choose Setup Connector.

  1. Open TypingMind in your browser.
  2. Click the Settings icon.
  3. Go to Advanced Settings.
  4. Open the Model Context Protocol section.
  5. Click Setup Connector and choose This Device.
TypingMind MCP connector setup screen with This Device selected
2

Run the connector command

Choose This Device, copy the command from TypingMind, and run it in Terminal. Keep the process running while you use MCP.

  1. Copy the setup command shown by TypingMind.
  2. Open Terminal on macOS or Windows Terminal on Windows.
  3. Paste and run the command.
  4. Approve the package install if Terminal asks you to proceed.
  5. Keep the Terminal window running while using MCP tools.
3

Add PromptX · AI应用原生专业能力增强系统 as a server

When the connector status is Ready, click Edit Servers and paste the MCP server configuration.

  1. Wait until the connector status shows Ready.
  2. Click Edit Servers.
  3. Paste the PromptX · AI应用原生专业能力增强系统 MCP server configuration.
  4. Save the server list.
  5. Refresh if you want to confirm the connector is still ready.
TypingMind MCP settings showing active server and Edit Servers button
{
  "mcpServers": {
    "deepractice-promptx": {
      "command": "npx",
      "args": [
        "-y",
        "root"
      ]
    }
  }
}
4

Use it across models

Save the server list, open Plugins, enable the PromptX · AI应用原生专业能力增强系统 MCP tools, then select any supported AI model in TypingMind and use the tools in chat or assign them to an AI agent.

  1. Open the Plugins page in TypingMind.
  2. Enable the PromptX · AI应用原生专业能力增强系统 MCP tools.
  3. Start a chat and choose the AI model you want to use.
  4. Use the MCP tools in chat or assign them to an AI agent.
  5. Switch to another AI model whenever needed without reconnecting MCP.
TypingMind chat using enabled MCP tools with a selected AI model
Can you use PromptX · AI应用原生专业能力增强系统 to help me with this task?
PromptX · AI应用原生专业能力增强系统
Sure. I read it.
Here is what I found using PromptX · AI应用原生专业能力增强系统.

Frequently asked questions

What is the PromptX · AI应用原生专业能力增强系统 MCP server used for?

PromptX · AI应用原生专业能力增强系统 is an MCP server that lets compatible AI clients connect to external tools and context. In TypingMind, you can add this MCP server once and make its tools available in your AI workspace.

Can I use PromptX · AI应用原生专业能力增强系统 MCP with multiple AI models in TypingMind?

Yes. TypingMind connects MCP tools at the workspace level, so you can use PromptX · AI应用原生专业能力增强系统 with different AI models such as Claude, ChatGPT, Gemini, or other models you have configured in TypingMind without setting up the MCP server separately for each model.

Why use PromptX · AI应用原生专业能力增强系统 MCP with TypingMind?

TypingMind is one of the best frontends for LLM chat because it brings multiple AI models, prompts, plugins, AI agents, API keys, and MCP tools into one workspace. With PromptX · AI应用原生专业能力增强系统 connected, you can use its MCP tools across your preferred models while keeping your chat workflow organized in TypingMind.

How do I connect PromptX · AI应用原生专业能力增强系统 MCP to TypingMind?

PromptX · AI应用原生专业能力增强系统 runs through the TypingMind local MCP connector. This is best when the MCP server needs access to local files, desktop apps, command-line tools, or private resources on your computer.

What tools does PromptX · AI应用原生专业能力增强系统 MCP provide in TypingMind?

PromptX · AI应用原生专业能力增强系统 exposes 7 MCP tools that can be enabled from the TypingMind Plugins page and used in chat or assigned to AI agents.

Do I need to share my API keys with TypingMind to use PromptX · AI应用原生专业能力增强系统 MCP?

No. TypingMind is local-first and lets you keep your model providers, API keys, prompts, and MCP configuration under your control. If PromptX · AI应用原生专业能力增强系统 requires authentication, add the required headers, OAuth settings, or local configuration for that MCP server when you create the connection.

Related MCP Servers

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇