Skip to main content

Documentation Index

Fetch the complete documentation index at: https://kindling.birklid.com/llms.txt

Use this file to discover all available pages before exploring further.

Open SourceApplications Layer

​
Agent Zero

A fully self-hosted agent framework that runs in Docker, modifies its own tools, and learns from experience through persistent vector memory.
Agent Zero GitHub

Type

Open Source (MIT)

Stack Layer

Applications

Language

Python

Stars

17.6k+

​
What it is

Agent Zero is an open-source autonomous agent framework built around a persistent, self-modifying Docker environment. It runs in a Linux container and gives agents full shell access, file editing capabilities, and long-term memory via FAISS vector search. Agents can spawn sub-agents in hierarchies, learn from completed tasks, and dynamically create or modify tools — including their own codebase. Speech interaction (TTS/STT) is built in, and it supports multiple LLM backends including OpenAI, Anthropic, and others. The project’s philosophy is radical transparency: the entire system is inspectable and community-owned, with no hidden cloud infrastructure or vendor lock-in. This makes it a popular starting point for researchers and developers who want to understand agent internals or build custom frameworks on top of a known baseline.
Use this when you want a fully self-hosted agent with no cloud dependency — especially for tasks requiring shell access, file manipulation, and memory that persists between sessions.

​
Get started

agent-zero.ai ↗

Architecture overview and documentation.

GitHub ↗

Source code, Docker setup, and community.

Ruflo

Multi-agent orchestration platform with 100+ specialized agents.

Symphony

OpenAI’s agent orchestration system for autonomous project work.