atomic-agent
Atomic Agent is a local-first AI agent. Runs open-weight models on your own machine via llama.cpp.
Atomic Agent is a local-first AI agent that runs on your own hardware using open-weight models via llama.cpp, eliminating per-token costs and ensuring privacy by default. It autonomously drives desktop workflows—browser automation, file editing, shell commands, document processing, and memory management—through a multi-step control loop optimized for small quantized models. The agent exposes integration points via HTTP API, Tauri sidecar, and Telegram, making it embeddable in desktop applications while maintaining full observability through NDJSON traces and approval gates.
Key Features
Use Cases
- 01Running autonomous coding tasks on consumer hardware without cloud API costs using local quantized models
- 02Automating web research and data extraction with browser-driven workflows and approval-gated actions
- 03Building desktop applications that embed AI agent capabilities via HTTP or Tauri sidecar integration
- 04Managing scheduled tasks (cron jobs, webhooks) with persistent memory across sessions
- 05Controlling agent workflows remotely via Telegram with inline approval buttons
- 06Integrating with existing Claude Code or OpenAI Codex subscriptions to drive flat-rate access through the agent loop
Related Agents
View morehermes-agent
The agent that grows with you
agency-agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
openinterpreter
A coding agent for open models like Kimi K3
cline
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
atomic-agent — FAQ
What is Atomic Agent?+
Atomic Agent is a local-first AI agent runtime that runs open-weight models on your machine via llama.cpp, executing multi-step workflows with browser automation, file operations, shell commands, and persistent memory. It operates as a standalone agent rather than a plugin for other assistants.
How do I install Atomic Agent?+
On macOS/Linux run 'curl -fsSL https://atomicagent.io/install | sh', or on Windows use 'irm https://atomicagent.io/install.ps1 | iex'. The installer downloads the CLI, support assets, and sets up PATH automatically.
Which AI clients does Atomic Agent work with?+
Atomic Agent is a standalone agent runtime, not a plugin. It can be embedded in custom applications via HTTP API or Tauri sidecar, and optionally driven by existing Claude Code or OpenAI Codex CLI subscriptions as completion providers.
What are the prerequisites and do I need API keys?+
Requires Node.js (dev only; releases are standalone binaries), a Chromium-family browser, and git. No API keys required for local operation with llama.cpp; optional cloud provider keys (OpenAI, OpenRouter, etc.) and service keys (GitHub, Telegram, Composio) can be configured.
Is Atomic Agent free to use?+
Yes, Atomic Agent is open-source with local execution incurring no per-token fees. Cloud provider costs apply only if you configure external LLM services instead of local models.
How does it handle dangerous operations like file writes or shell commands?+
All risky actions route through an approval ladder with interactive prompts, configurable permission modes (plan/default/auto/bypass), and session-based grants. Every decision and tool execution is logged to append-only NDJSON traces for full observability.
How do I install atomic-agent?+
Open the source repository on GitHub and follow its README. atomic-agent is a agent — MCP Agents Market links you directly to the official repo.
Is atomic-agent free?+
atomic-agent is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.