DEV.co
MCP Servers · heshengtao

super-agent-party

Super Agent Party is a self-hosted AI desktop companion framework supporting VRM/Live2D avatars, multi-modal chat, computer vision-based task automation, and deployment to messaging/streaming platforms. Built in JavaScript, it enables conversational AI agents with extensible skills, MCP integration, and browser automation.

Source: GitHub — github.com/heshengtao/super-agent-party
2.5k
GitHub stars
255
Forks
JavaScript
Primary language
AGPL-3.0
License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

FieldValue
Repositoryheshengtao/super-agent-party
Ownerheshengtao
Primary languageJavaScript
LicenseAGPL-3.0 — OSI-approved
Stars2.5k
Forks255
Open issues16
Latest releasev0.4.2 (2026-06-21)
Last updated2026-07-08
Sourcehttps://github.com/heshengtao/super-agent-party

What super-agent-party is

Node.js/Electron-based desktop application providing an agentic framework with desktop vision, keyboard/mouse control, extensible agent skills, MCP protocol support, and OpenAI-compatible API surface. Supports multi-agent group chat, long-term memory, and platform-specific bots (Discord, Telegram, Bilibili, YouTube, etc.).

Quickstart

Get the super-agent-party source

Clone the repository and explore it locally.

terminalbash
git clone https://github.com/heshengtao/super-agent-party.gitcd super-agent-party# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

AI-Powered Desktop Automation & Task Execution

Agents can autonomously execute computer tasks (filling forms, data entry, testing) via vision + keyboard/mouse control, reducing manual workflow overhead for repetitive operations.

Multi-Platform Conversational Bot Deployment

Single codebase deployment to Discord, Telegram, WeChat, QQ, Bilibili, YouTube, and Twitch without reimplementation, reducing platform-specific development cost.

Custom AI Companion with Extensible Skills

Developer teams building branded AI assistants with custom VRM avatars, real-time chat, agent skills bootstrapping, and browser integration for information retrieval workflows.

Implementation considerations

  • AGPL-3.0 copyleft: any internal modifications trigger open-source obligation; clarify IP strategy with legal before extending.
  • Ecosystem maturity: v0.4.2 is recent (June 2026); production readiness and API stability require testing; active development suggests breaking changes possible.
  • Agent skill bootstrapping requires JavaScript/Node.js competency; extension system exists but documentation depth for complex skills unknown.
  • Desktop vision + automation relies on OS-level capture APIs (Windows 10/11, macOS M-chip); cross-platform stability and permission model requires validation.
  • Multi-LLM support inferred (Claude, OpenAI references in description); underlying model integration mechanism and fallback behavior not detailed.

When to avoid it — and what to weigh

  • Strict Commercial License Requirements — AGPL-3.0 requires derivative works to be open-source and triggers copyleft obligations on modification. Organizations unable or unwilling to open-source extensions/modifications should avoid or require legal review.
  • Enterprise SLA & Support Expectations — Community-driven project with no documented commercial support, SLAs, or guaranteed response times. High-risk for mission-critical deployments requiring vendor accountability.
  • Lightweight/Edge-Device Deployment — Electron/Node.js stack is resource-heavy. Unsuitable for embedded systems, IoT, or low-power environments where footprint and latency are critical constraints.
  • Regulated Industries (Healthcare, Finance, Legal) — No evidence of security audits, data residency guarantees, or compliance certifications (HIPAA, SOC2, etc.). Risk profile undefined for regulated workloads.

License & commercial use

AGPL-3.0 (GNU Affero General Public License v3.0). Copyleft license: any modifications or network deployment trigger requirement to disclose source and license derivative works identically. Commercial use is permitted only if all modifications are open-sourced.

Commercial use is legally permissible under AGPL-3.0, but with mandatory conditions: (1) any modifications must be released under AGPL-3.0, (2) source code must be provided to users, and (3) internal SaaS/hosted modifications must disclose source on request. **Requires legal review** before embedding in proprietary products or closed-source extensions. Consult IP counsel if commercial viability depends on trade secrets or closed-source derivative works.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

SignalAssessment
MaintenanceActive
DocumentationAdequate
License clarityNeeds review
Deployment complexityModerate
DEV.co fitGood
Assessment confidenceMedium
Security considerations

Desktop agent with keyboard/mouse/vision automation inherits OS privilege risks (e.g., UAC bypass, screen capture scope). No documented threat model, input sanitization strategy, or code review process. OpenAI-compatible API surface requires API key management; local storage and transmission encryption not specified. VRM avatar uploads and external extension installation create supply-chain risk vectors. Security audit status: unknown.

Alternatives to consider

LangChain / LangGraph (Python)

Framework-agnostic agent orchestration; mature, battle-tested, larger ecosystem; permissive MIT license; weaker desktop/avatar UI layer but stronger enterprise tooling.

n8n (Permissive Community Edition or Proprietary Cloud)

Visual workflow automation with LLM agents; built-in platform integrations (Discord, Telegram, etc.); no-code focus; weak desktop companion/avatar support; clearer commercial licensing options.

Dify (Apache 2.0)

Open-source LLM ops platform with agent workflows, RAG, and API deployment; permissive license; better compliance/audit posture; less desktop/avatar-native; stronger enterprise feature set.

Software development agency

Build on super-agent-party with DEV.co software developers

Super Agent Party offers extensible agent orchestration and multi-platform deployment—but AGPL-3.0 copyleft and evolving stability require careful planning. Let us help you evaluate licensing, architecture, and integration fit.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

Related on DEV.co

Explore the category and the services that help you build with it.

super-agent-party FAQ

Can we use this in a closed-source product?
Not without significant restrictions. AGPL-3.0 requires any modifications to be released under the same license. You may use unmodified binaries for SaaS, but source changes are copyleft-bound. **Legal review required.**
Does it support headless/backend-only operation?
Unclear. Desktop vision, avatar rendering, and chat UI are core; agent task execution and API endpoints may support headless modes, but documentation and production-ready patterns are not evident.
How do we add custom agent skills?
Extension system exists (agent skills, MCP integration mentioned); JavaScript/Node.js required; bootstrapping example (sap-extension-creator) noted but full API reference not provided in excerpt.
What LLMs are supported?
References to Claude, OpenAI, and MCP suggest multi-model support; exact model matrix, fallback behavior, and cost/latency characteristics unknown. Review documentation or source code.

Software development & web development with DEV.co

DEV.co helps companies turn open-source tools like super-agent-party into production software. Our software development services cover the full lifecycle — architecture, web development, integration, and maintenance — delivered by software developers and web developers who ship. Engage our software development agency to implement or customize it for your mcp servers stack.

Ready to Deploy Custom AI Agents?

Super Agent Party offers extensible agent orchestration and multi-platform deployment—but AGPL-3.0 copyleft and evolving stability require careful planning. Let us help you evaluate licensing, architecture, and integration fit.