by DemonDamon
A unified, production‑ready multi‑agent platform offering a Python SDK, CLI, Studio server, and desktop application, supporting meta‑agent orchestration, over 15 LLM providers, hierarchical memory, avatar/group chat, skill ecosystem, safety sandbox, and IM gateway.
AgenticX provides a complete stack for building, deploying, and operating collaborative AI agents. It combines a Python SDK, a command‑line interface (agx), a FastAPI‑based Studio server, and an Electron‑powered desktop client to let developers create anything from simple assistants to complex, multi‑agent enterprises.
pip install agenticx
Optional extras add memory, document processing, graph, monitoring, MCP protocol, etc.agx project create my-agent --template basic
agx serve --port 8000
from agenticx import Agent, Task, AgentExecutor
from agenticx.llms import OpenAIProvider
agent = Agent(id="data-analyst", name="Data Analyst", role="Data Analysis Expert")
task = Task(id="analysis-task", description="Analyze sales data trends")
llm = OpenAIProvider(model="gpt-4")
result = AgentExecutor(agent=agent, llm=llm).run(task)
print(result)
@tool decorator; agents will invoke them automatically.Q: Do I need to install heavy libraries like torch?
A: No. The core package is lightweight (~27 deps). Heavy libraries are optional extras (e.g., agenticx[ocr] pulls in torch).
Q: How can I add a new LLM provider? A: Implement a subclass of the provider interface and register it via the SDK; the built‑in routing will automatically include it.
Q: Is the platform production‑ready? A: Yes. It includes safety sandboxes, multi‑tenant session isolation, observability, and failsafe orchestration.
Q: Can I run AgenticX on Windows/macOS/Linux? A: Supported on all three; the desktop app is cross‑platform, and the server runs on any Python 3.10+ environment.
Q: How do I enable monitoring with Prometheus?
A: Install the monitoring extra (pip install "agenticx[monitoring]") and configure the Prometheus exporter; metrics are exposed via the Studio server.
Architecture • Features • Quick Start • Examples • Progress
LiteLLM (PyPI): Malicious releases litellm 1.82.7 and 1.82.8 were removed from PyPI after reports that they could exfiltrate API keys. If you ever installed either version, uninstall them, rotate any credentials that may have been exposed, and upgrade to a release the upstream project and PyPI list as safe (for example 1.82.9+, per current upstream guidance). Check your environment with pip show litellm.
AgenticX aims to create a unified, scalable, production-ready multi-agent application development framework, empowering developers to build everything from simple automation assistants to complex collaborative intelligent agent systems.
The framework is organized into 5 tiers: User Interface (Desktop / CLI / SDK) → Studio Runtime (Session Manager, Meta-Agent, Team Manager, Avatar & Group Chat) → Core Framework (Orchestration, Execution, Agent, Memory, Tools, LLM Providers, Hooks) → Platform Services (Observability, Protocols, Security, Storage) → Domain Extensions (GUI Agent, Knowledge & GraphRAG, AgentKit Integration).
ainvoke/astream) with a typed AgentEvent stream, multi-turn history in/out, parallel tool execution, and optional loop-detector / compactor / offloader injection — zero Studio/CLI coupling (legacy text-JSON TextReActAgent facade kept for compatibility)Offloader protocol + filesystem-backed FileOffloader keep large tool results / compressed context out of live history (inline reference placeholders, retrieved on demand); plus an in-workspace MCP gateway that runs MCP servers inside the sandbox.changelog versioning, source tagging, and per-skill enable/disableagx): serve, studio, loop, run, project, deploy, codegen, docs, skills, hooks, debug, scaffold, and config management# Core install (lightweight, no torch, installs in seconds)
pip install agenticx
# Install optional features as needed
pip install "agenticx[memory]" # Memory: mem0, chromadb, qdrant, redis, milvus
pip install "agenticx[document]" # Document processing: PDF, Word, PPT parsing
pip install "agenticx[graph]" # Knowledge graph: networkx, neo4j, community detection
pip install "agenticx[llm]" # Extra LLMs: anthropic, ollama
pip install "agenticx[monitoring]" # Observability: prometheus, opentelemetry
pip install "agenticx[mcp]" # MCP protocol
pip install "agenticx[database]" # Database backends: postgres, SQLAlchemy
pip install "agenticx[data]" # Data analysis: pandas, scikit-learn, matplotlib
pip install "agenticx[ocr]" # OCR (pulls in torch ~2GB): easyocr
pip install "agenticx[volcengine]" # Volcengine AgentKit
pip install "agenticx[all]" # Everything
Tip: The core package includes only ~27 lightweight dependencies and installs in seconds. Heavy dependencies (torch, pandas, etc.) are optional extras - install only what you need.
Browser automation: To run browser-use as an MCP server from AgenticX (
mcp_connect/mcp_call), see examples/browser-use-mcp.md.
Desktop MCP upgrades (2026-04): Near Settings now supports MCP brand auto-discovery (Cursor / Trae / Claude / OpenClaw / Hermes / Codex), built-in Monaco JSON editor with schema validation, and one-click install from ModelScope MCP marketplace.
# Clone repository
git clone https://github.com/DemonDamon/AgenticX.git
cd AgenticX
# Using uv (recommended, 10-100x faster than pip)
pip install uv
uv pip install -e . # Core install
uv pip install -e ".[memory,graph]" # Add optional features
uv pip install -e ".[all]" # Everything
uv pip install -e ".[dev]" # Development tools
# Or using pip
pip install -e .
pip install -e ".[all]"
# Set environment variables
export OPENAI_API_KEY="your-api-key"
export ANTHROPIC_API_KEY="your-api-key" # Optional
Complete Installation Guide: For system dependencies (antiword, tesseract) and advanced document processing features, see INSTALL.md
After installation, the agx command-line tool is available:
# View version
agx --version
# Create a new project
agx project create my-agent --template basic
# Start the API server
agx serve --port 8000
# Parse documents (PDF/PPT/Word etc.)
agx mineru parse report.pdf --output ./parsed
Full CLI Reference: See docs/cli.md for complete command documentation.
from agenticx import Agent, Task, AgentExecutor
from agenticx.llms import OpenAIProvider
# Create agent
agent = Agent(
id="data-analyst",
name="Data Analyst",
role="Data Analysis Expert",
goal="Help users analyze and understand data",
organization_id="my-org"
)
# Create task
task = Task(
id="analysis-task",
description="Analyze sales data trends",
expected_output="Detailed analysis report"
)
# Configure LLM
llm = OpenAIProvider(model="gpt-4")
# Execute task
executor = AgentExecutor(agent=agent, llm=llm)
result = executor.run(task)
print(result)
from agenticx.tools import tool
@tool
def calculate_sum(x: int, y: int) -> int:
"""Calculate the sum of two numbers"""
return x + y
@tool
def search_web(query: str) -> str:
"""Search web information"""
return f"Search results: {query}"
# Agents will automatically invoke these tools
We provide rich examples demonstrating various framework capabilities:
Single Agent Example
# Basic agent usage
python examples/m5_agent_demo.py
Multi-Agent Collaboration
# Multi-agent collaboration example
python examples/m5_multi_agent_demo.py
Simple Workflow
# Basic workflow orchestration
python examples/m6_m7_simple_demo.py
Complex Workflow
# Complex workflow orchestration
python examples/m6_m7_comprehensive_demo.py
A2A Protocol Demo
# Inter-agent communication protocol
python examples/m8_a2a_demo.py
Complete Monitoring Demo
# Observability module demo
python examples/m9_observability_demo.py
Basic Memory Usage
# Memory system example
python examples/memory_example.py
Healthcare Scenario
# Healthcare memory scenario
python examples/mem0_healthcare_example.py
Human Intervention Flow
# Human-in-the-loop example
python examples/human_in_the_loop_example.py
Detailed documentation: examples/README_HITL.md
Chatbot
# LLM chat example
python examples/llm_chat_example.py
Code Execution Sandbox
# Micro-sandbox example
python examples/microsandbox_example.py
Technical blog: examples/microsandbox_blog.md
Intelligent Intent Recognition System
# Intent recognition service example
python examples/agenticx-for-intent-recognition/main.py
A production-grade, layered intent recognition service built entirely on the AgenticX framework, demonstrating real-world usage of Agents, Workflows, Tools, and Storage systems.
Architecture:
IntentRecognitionAgent (LLM-powered) with specialized agents (GeneralIntentAgent, SearchIntentAgent, FunctionIntentAgent) for fine-grained classificationUIE + LLM + Rule extractors with confidence-weighted fusion), regex/full-text matching, and a full post-processing suite (confidence adjustment, conflict resolution, entity optimization, intent refinement)UnifiedStorageManagerKey capabilities:
See: examples/agenticx-for-intent-recognition/
GUI Automation Agent
# GUI Agent example
python examples/agenticx-for-guiagent/AgenticX-GUIAgent/main.py
Key capabilities:
See: examples/agenticx-for-guiagent/
| Project | Description | Path |
|---|---|---|
| Agent Skills | Skill discovery, matching, and SOP-driven skill execution for agents | examples/agenticx-for-agent-skills/ |
| AgentKit | Volcengine AgentKit integration with Docker-ready agent deployment | examples/agenticx-for-agentkit/ |
| ChatBI | Conversational BI — natural language to data insights | examples/agenticx-for-chatbi/ |
| Deep Research | Multi-source deep research and report generation | examples/agenticx-for-deepresearch/ |
| Doc Parser | Intelligent document parsing (PDF, Word, PPT) | examples/agenticx-for-docparser/ |
| Finance | Financial news hunting and analysis | examples/agenticx-for-finance/ |
| Future Prediction | Predictive analysis and forecasting | examples/agenticx-for-future-prediction/ |
| GraphRAG | Knowledge graph-enhanced retrieval-augmented generation | examples/agenticx-for-graphrag/ |
| Math Modeling | Mathematical modeling assistant | examples/agenticx-for-math-modeling/ |
| Model Architecture Discovery | Automated model architecture search and discovery | examples/agenticx-for-modelarch-discovery/ |
| Query Optimizer | SQL/query optimization agent | examples/agenticx-for-queryoptimizer/ |
| Sandbox | Secure code execution sandbox | examples/agenticx-for-sandbox/ |
| Spec Coding | Specification-driven code generation | examples/agenticx-for-spec-coding/ |
| Vibe Coding | AI-assisted creative/vibe coding | examples/agenticx-for-vibecoding/ |
| Module | Status | Description |
|---|---|---|
| M1 | ✅ | Core Abstraction Layer — Agent, Task, Tool, Workflow, Event Bus, Component, and Pydantic data contracts |
| M2 | ✅ | LLM Service Layer — 15+ providers (OpenAI / Anthropic / Ollama / Gemini / Kimi / MiniMax / Ark / Zhipu / Qianfan / Bailian), response caching, failover routing |
| M3 | ✅ | Tool System — Function decorators, MCP Hub, remote tools v2, OpenAPI toolset, sandbox tools, skill bundles, document routers |
| M4 | ✅ | Memory System — Hierarchical (core / episodic / semantic), Mem0, workspace, short-term, memory decay, hybrid search, memory intelligence engine |
| M5 | ✅ | Agent Core — Meta-Agent CEO dispatcher, think-act loop, event-driven architecture, self-repair, overflow recovery, reflection |
| M6 | ✅ | Task Validation — Pydantic-based output parsing, auto-repair, guiderails |
| M7 | ✅ | Orchestration Engine — Graph-based workflow engine + Flow system with decorators, execution plans, conditional routing, parallel execution |
| M8 | ✅ | Communication Protocols — A2A (client / server / AgentCard / skill-as-tool), MCP resource access, AGUI protocol |
| M9 | ✅ | Observability — Callbacks, real-time monitoring, trajectory analysis, span tree, WebSocket streaming, Prometheus / OpenTelemetry integration |
| M10 | ✅ | Developer Experience — CLI (agx with 15+ commands), Studio Server (FastAPI), Desktop App (Electron + React + Zustand, Pro/Lite dual mode) |
| M11 | ✅ | Enterprise Security — Safety layer (leak detection / sanitizer / injection detector / policy / audit), Sandbox (Docker / Microsandbox / Subprocess / Jupyter kernel / code interpreter) |
| M13 | ✅ | Knowledge & Retrieval — Knowledge base with document processing, chunkers, graphers (GraphRAG), readers; retrieval (vector / BM25 / graph / hybrid / auto); embeddings (OpenAI / Bailian / SiliconFlow / LiteLLM) |
| M14 | ✅ | Avatar & Collaboration — Avatar registry, group chat (user-directed / meta-routed / round-robin), delegation, role-playing, conversation patterns, team management |
| M15 | ✅ | Evaluation Framework — EvalSet, LLM judge, composite judge, span evaluator, trajectory matcher, trace converter |
| M16 | ✅ | Embodiment — GUI Agent framework with action reflection, stuck detection, action caching, REACT parsing, device-cloud routing, DAG verification, human-in-the-loop |
| M17 | ✅ | Storage Layer — Key-Value (SQLite / Redis / PostgreSQL / MongoDB), Vector (Milvus / Qdrant / Chroma / Faiss / PgVector / Pinecone / Weaviate), Graph (Neo4j / Nebula), Object (S3 / GCS / Azure) |
| Module | Status | Description |
|---|---|---|
| M12 | 🚧 | Agent Evolution — Architecture search, knowledge distillation, adaptive planning |
| M18 | 🚧 | Multi-tenancy & RBAC — Session-level tenant_id isolation landed; fine-grained permission control in progress |
| Capability | Status | Description |
|---|---|---|
| Skill Self-Evolution | ✅ | Runtime tool-call observation capture, auto-skill creation via session review, quality gate / usage stats / deprecation loop (learning) |
| Multi-Brain Knowledge | ✅ | Isolatable/mountable "doc brain + code brain" with cross-brain search (brain) + multi-codebase hybrid semantic index (code_index) |
| Long-Horizon Coding | ✅ | Long-run orchestration (multi-source / isolated workspaces / stall self-healing / continuation backoff, longrun) + disk-backed project state machine (project_state) |
| IM Channel Integration | ✅ | Remote command gateway for Feishu / WeCom / DingTalk / personal WeChat (iLink) (gateway) |
| Claude Code Bridge | ✅ | Token-protected local HTTP/NDJSON control plane, headless / visible TUI dual mode (cc_bridge) |
| Extension Ecosystem | ✅ | AGX Bundle definitions, local install/uninstall, multi-source registry aggregated search (extensions) |
| Embeddable ReActAgent | ✅ | Canonical async function-calling ReAct SDK primitive with typed event stream, multi-turn history, parallel tools, optional loop-detector / compactor / offloader, zero Studio/CLI coupling (agents) |
| Unified Offload & MCP Gateway | ✅ | Offloader protocol + FileOffloader for out-of-history large payloads, plus in-workspace MCP gateway (AgentScope v2 P0 internalization, core.offload / sandbox.mcp_gateway) |
pyproject.toml)pip install "agenticx[xxx]"We welcome community contributions! Please refer to:
The personal WeChat (iLink) channel integration in AgenticX was built on top of the openilink-sdk-go library from OpeniLink Hub. We specifically relied on:
FetchQRCode / PollQRStatus APIs for the scan-to-bind UXclient.Monitor() for real-time inbound message streamingSendText / Push for reply delivery with context_token routingDownloadMedia / DownloadVoice for encrypted WeChat mediaOpeniLink Hub's OpenClaw App also demonstrated an AI Agent gateway integration pattern that informed our adapter architecture.
We did not include OpeniLink Hub's web console, App Marketplace, or multi-bot management features. AgenticX's core multi-agent runtime, session management, and Desktop UI remain fully independent implementations.
OpeniLink Hub — MIT License — github.com/openilink/openilink-hub
Additional reference: WorkBuddy — WeixinBot Guide for iLink protocol usage patterns.
Desktop development: The iLink Go sidecar binary is not committed to this repository. Before using the personal WeChat bridge locally, run make build in packaging/wechat-sidecar/ (requires Go 1.22+). See packaging/wechat-sidecar/README.md.
This project is licensed under the Apache License, Version 2.0 — see the LICENSE file for details.
AgenticX would not exist in its current form without the inspiration, architectural ideas, and engineering wisdom we drew from the open-source community. We have studied the following projects in depth, and we are genuinely grateful to every author, contributor, and community behind them.
| Project | Repository | What we learned |
|---|---|---|
| A2A | a2aproject/A2A | Agent-to-Agent protocol design |
| AgentCPM-GUI | OpenBMB/AgentCPM-GUI | Compact GUI action schema & RFT training |
| ADK Python | google/adk-python | Agent lifecycle, runner abstractions |
| ag-ui | ag-ui-protocol/ag-ui | Agent–UI streaming protocol |
| AgentKit SDK | volcengine/agentkit-sdk-python | Agent deployment & skill packaging |
| AgentRun SDK | Serverless-Devs/agentrun-sdk-python | Serverless agent runtime patterns |
| AgentScope | agentscope-ai/agentscope | Multi-agent communication & pipeline |
| Agno | agno-agi/agno | Lightweight agent framework design |
| Camel | camel-ai/camel | Role-playing agents & society simulation |
| Cherry Studio | CherryHQ/cherry-studio | Desktop UX, MCP integration, skill system |
| Claude Code | anthropics/claude-code | Agentic CLI UX & plugin architecture |
| CLI-Anything | HKUDS/CLI-Anything | CLI-native agent harness |
| ClawTeam | HKUDS/ClawTeam | Multi-agent team coordination |
| CodexMonitor | Dimillian/CodexMonitor | Desktop monitoring & Tauri app patterns |
| CrewAI | crewAIInc/crewAI | Crew orchestration, flow & memory system |
| DeepWiki Open | AsyncFuncAI/deepwiki-open | Repository-level knowledge indexing |
| Deer Flow | bytedance/deer-flow | Deep research workflow & skill harness |
| Eigent | eigent-ai/eigent | Multi-agent workforce & SSE event spec |
| Iron Claw | nearai/ironclaw | Agent evaluation & benchmark harness |
| JoyAgent / JD Genie | jd-opensource/joyagent-jdgenie | Enterprise agent orchestration |
| Khazix Skills | KKKKhazix/Khazix-Skills | Skill module structure & packaging |
| Lobe Icons | lobehub/lobe-icons | AI provider icon design system |
| LoongSuite Python Agent | alibaba/loongsuite-python-agent | OpenTelemetry GenAI instrumentation |
| MAI-UI | Tongyi-MAI/MAI-UI | Device-cloud collaboration & GUI grounding |
| Microsandbox | zerocore-ai/microsandbox | Lightweight sandboxed code execution |
| MobiAgent | IPADS-SAI/MobiAgent | Mobile multi-stage planning |
| MobileAgent | X-PLUG/MobileAgent | Multi-agent mobile GUI automation |
| Model Context Protocol | modelcontextprotocol/modelcontextprotocol | Standardized LLM tool/resource protocol |
| NVIDIA NemoClaw | NVIDIA/NemoClaw | GPU-accelerated agent plugin system |
| OpenClaw | openclaw/openclaw | Open desktop agent platform & extensions |
| OpenSandbox | alibaba/OpenSandbox | Container-based code sandbox |
| OpenShell | NVIDIA/OpenShell | Rust-based secure agent shell |
| OpenSkills | numman-ali/openskills | Skill registry & discovery |
| OWL | camel-ai/owl | Embodied multi-agent collaboration |
| Pydantic AI | pydantic/pydantic-ai | Type-safe agent & eval framework |
| Refly | refly-ai/refly | AI-native knowledge canvas UX |
| Serverless Devs | Serverless-Devs/Serverless-Devs | Serverless agent deployment toolchain |
| Skills | anthropics/skills | Skill definition format & lifecycle |
| Spring AI | spring-projects/spring-ai | Enterprise AI abstraction patterns |
| SWE-agent | SWE-agent/SWE-agent | Software engineering agent & ACR loop |
| VE ADK | volcengine/veadk-python | Skills system & cloud-native A2A |
| ZeroBoot | zerobootdev/zeroboot | Zero-config agent bootstrapping |
Thank you for building in the open. Your work has been a constant source of insight and motivation for the AgenticX team.
If AgenticX helps you, please give us a Star!
Please log in to share your review and rating for this MCP.
Explore related MCPs that share similar capabilities and solve comparable challenges
by modelcontextprotocol
An MCP server implementation that provides a tool for dynamic and reflective problem-solving through a structured thinking process.
by danny-avila
Provides a self‑hosted ChatGPT‑style interface supporting numerous AI models, agents, code interpreter, image generation, multimodal interactions, and secure multi‑user authentication.
by block
Automates engineering tasks on local machines, executing code, building projects, debugging, orchestrating workflows, and interacting with external APIs using any LLM.
by RooCodeInc
Provides an autonomous AI coding partner inside the editor that can understand natural language, manipulate files, run commands, browse the web, and be customized via modes and instructions.
by pydantic
A Python framework that enables seamless integration of Pydantic validation with large language models, providing type‑safe agent construction, dependency injection, and structured output handling.
by mcp-use
A Python SDK that simplifies interaction with MCP servers and enables developers to create custom agents with tool‑calling capabilities.
by lastmile-ai
Build effective agents using Model Context Protocol and simple, composable workflow patterns.
by Klavis-AI
Provides production‑ready MCP servers and a hosted service for integrating AI applications with over 50 third‑party services via standardized APIs, OAuth, and easy Docker or hosted deployment.
by nanbingxyz
A cross‑platform desktop AI assistant that connects to major LLM providers, supports a local knowledge base, and enables tool integration via MCP servers.