atomic-agent

Atomic Agent is an open-source AI assistant that is locally focused and works on your own hardware. When managing the control cycle and the entire situation locally, it offers the ability to control the browser, edit files, and execute approved shell commands. Thanks to Llama.cpp support, it remembers context between sessions using local or cloud models.

AT

AtomicBot-ai · Local AI / Self-Hosted LLM

STARS3.5K
FORKS276
COMMITS0
CONTRIBUTORS0
VERSIONv0.6.11
UPDATED2 days ago
TypeScriptAI Agentllama.cppLocal-FirstMIT

Common commercial products in this category: ChatGPT, Claude, Gemini

What can it do?
  • You can browse the browser, browse the documents, and read and edit the files.
  • You can execute approved shell commands and call external tools via MCP.
  • With Fusion mode, you can distribute planning and execution tasks to local or cloud models.
User reviews
  • ↑ PositiveGitHub Issue

    Feature request: Hindsight as memory implementation

    “It would be a great improvement to have Hindsight as memory provider since it's state of the art and fully local. Oh My Pi uses it a memory backend for example I can help with the integration”

  • ↓ NegativeGitHub Issue

    Feature: User-configurable shell guard rules via config.json + /guard menu command

    “# Feature: User-configurable shell guard rules via config.json + `/guard` menu command ## Problem When running atomic-agent via Telegram (or any non-interactive surface), every shell command that doesn't match a built-in guard rule triggers an approval prompt. This creates a…”

My note

Alternative pages

More in: Local AI / Self-Hosted LLM