LLM.coPrivate, self-hosted LLMs & custom AI applications.
Automatic.coAI-powered workflow & business process automation.
RMA.aiAI-powered cybersecurity monitoring, threat detection & defense.
VB.coOperating AI™ — one intelligent layer for middle-market ops.
LAW.co
RFP.co
Search.co
VDR.aiAstrBot
AstrBot is an open-source Python-based AI agent platform that integrates with messaging apps (QQ, Telegram, WeChat, Slack, etc.) and LLMs to build conversational AI. It supports plugins, knowledge bases, and agent frameworks via a web UI and offers sandbox execution for safe code/shell operations.
Key facts
Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.
| Field | Value |
|---|---|
| Repository | AstrBotDevs/AstrBot |
| Owner | AstrBotDevs |
| Primary language | Python |
| License | AGPL-3.0 — OSI-approved |
| Stars | 36k |
| Forks | 2.5k |
| Open issues | 1.3k |
| Latest release | v4.26.5 (2026-07-07) |
| Last updated | 2026-07-07 |
| Source | https://github.com/AstrBotDevs/AstrBot |
What AstrBot is
Python 3.12+ framework providing multi-platform IM connectors, LLM orchestration (OpenAI, Gemini, Llama, etc.), MCP support, plugin architecture (1000+ available), agent sandbox for isolated execution, and auto-context compression. Deployable via Docker, uv, or cloud platforms.
Get the AstrBot source
Clone the repository and explore it locally.
git clone https://github.com/AstrBotDevs/AstrBot.gitcd AstrBot# follow the project's README for install & configurationNeed it deployed, integrated, or customized instead? DEV.co ships production installs.
Best use cases
Implementation considerations
- AGPL-3.0 license: any modifications or derivative deployments must be open-sourced; review legal/compliance before use in proprietary products.
- 1,298 open issues and high fork count suggest active development but potential stability gaps; test thoroughly on target IM platforms before production.
- Requires uv (package manager) or Docker for deployment; ensure team familiarity with Python 3.12+, async I/O, and plugin lifecycle management.
- LLM integration is by-bring-your-own-key (OpenAI, Gemini, etc.); no bundled model; cost scales with API usage and context length.
- Sandbox execution adds complexity; validate resource limits, timeout handling, and isolation for untrusted code execution scenarios.
When to avoid it — and what to weigh
- Closed-Source or Proprietary Requirement — AGPL-3.0 licensing requires source code disclosure and derivative works to remain open. If you cannot publish modifications or need permissive commercial licensing, this is unsuitable.
- High Security/Compliance Constraints — No explicit audit trail, SOC 2, or HIPAA compliance mentioned. Code/shell sandbox execution presents attack surface; requires threat modeling for regulated data (healthcare, finance).
- Single-Platform, Lightweight Requirement — Overhead of multi-platform abstraction, plugin system, and agent orchestration may be excessive for a simple single-channel bot. Consider a purpose-built lightweight library instead.
- Limited Python Ecosystem in Your Stack — Python-only implementation; integration with JVM, Go, or .NET applications requires API/RPC bridges, adding operational complexity.
License & commercial use
AGPL-3.0 (GNU Affero General Public License v3.0). This is a strong copyleft license. Any modifications or network-distributed versions must disclose source and remain open-source. Use only with explicit legal review if commercial, proprietary, or embedded in closed-source products.
AGPL-3.0 permits use but requires: (1) all modifications be open-sourced; (2) if deployed as a service, source code must be offered to users. Commercial support/licensing from maintainers is not described in provided data. **Requires review with your legal team before commercial deployment.**
DEV.co evaluation signals
Editorial assessment — not user reviews. Directional, with an explicit confidence level.
| Signal | Assessment |
|---|---|
| Maintenance | Active |
| Documentation | Adequate |
| License clarity | Clear |
| Deployment complexity | Moderate |
| DEV.co fit | Good |
| Assessment confidence | High |
Agent sandbox for code/shell execution is a security feature but also an attack surface. No mention of input validation, prompt injection mitigations, or LLM jailbreak protections. Plugin ecosystem introduces supply-chain risk (1000+ plugins of unknown provenance/maintenance). Code execution from untrusted sources (e.g., user-provided agents) requires strict isolation, resource limits, and monitoring. Assess threat model against LLM output, plugin code, and user input before production.
Alternatives to consider
Dify
Closed-UI agent/workflow builder; emphasizes enterprise SaaS deployment and visual orchestration over self-hosting; may require licensing.
LangChain / LangGraph
Python LLM framework focused on agent chains and tool use; more modular and lightweight but lacks built-in IM integrations and requires custom platform adapters.
n8n
Visual no-code workflow automation with LLM support; easier onboarding for non-technical users but less customizable for agent-specific logic.
Build on AstrBot with DEV.co software developers
Review the full documentation at astrbot.app, assess AGPL-3.0 licensing implications with your legal team, and deploy via Docker or uv. Start with a non-production test on your IM platform.
Talk to DEV.coRelated on DEV.co
Explore the category and the services that help you build with it.
AstrBot FAQ
Can I use AstrBot commercially?
Does AstrBot include LLMs?
How stable is AstrBot for production?
Is AstrBot secure for user data?
Custom software development services
DEV.co builds and maintains ai frameworks systems for startups through enterprises. If AstrBot fits your architecture, our custom software development services cover implementation, migration, and the long tail of maintenance.
Ready to Deploy Multi-Platform AI?
Review the full documentation at astrbot.app, assess AGPL-3.0 licensing implications with your legal team, and deploy via Docker or uv. Start with a non-production test on your IM platform.