Skip to content
Sign in
  • Browse all toolsFilter by category and pricing
  • Browse categoriesWriting, code, image, audio and more
  • Compare toolsPut up to three side by side
  • Read the blogGuides and comparisons

IronClaw

Open-source personal AI assistant with local storage, sandboxed tools, scheduled automation, and MCP support.

IronClaw is a personal AI assistant and open-source Agent OS that runs sandboxed tools, automates background tasks, and maintains searchable memory. IronClaw serves users and developers seeking local data storage and extensibility. IronClaw is available under Apache or MIT licensing through downloadable installers or source builds, with LLM provider setup required.

Updated 9 Nov 20266 sections

IronClaw website screenshot

What is IronClaw?

IronClaw is an open-source personal AI assistant and Agent OS focused on privacy, security, and extensibility. IronClaw is written in Rust and described as a reimplementation inspired by OpenClaw. The project supports personal and professional assistant workflows while keeping application state locally and providing auditable source code.

IronClaw combines a conversational interface with tools, persistent memory, and background task execution. Users can interact through a browser, terminal, HTTP webhooks, or supported messaging channels. The browser gateway includes chat, memory, jobs, logs, extensions, and routines, bringing these functions into a shared interface.

What IronClaw can do

IronClaw supports interactive requests and ongoing automation through an agent loop, scheduler, routines engine, and tool registry. Security controls cover untrusted tool execution, credential handling, and external content. Extension options include MCP servers and WASM plugins, while persistent memory combines full-text and vector search.

  • IronClaw runs untrusted tools in isolated WebAssembly containers with explicit permissions.
  • IronClaw schedules routines using cron, events, and webhooks.
  • IronClaw builds WASM tools from descriptions of required capabilities.
  • IronClaw connects to Model Context Protocol servers for additional tools.
  • IronClaw searches persistent memory using full-text and vector retrieval.
  • IronClaw handles parallel requests with isolated contexts.
  • IronClaw detects and attempts recovery from stuck operations.
  • IronClaw provides Docker sandbox execution with per-job tokens.

How IronClaw works

IronClaw installation begins with a selected release from the repository’s Releases page or a source build. The ironclaw onboard command guides users through choosing an LLM provider, entering an API key in a hidden prompt, and selecting a model. Onboarding provisions local configuration, an encrypted credential store, and a WebUI login token.

IronClaw installs and starts a background service during onboarding on macOS and Linux, then prints a WebUI link. Windows users can start the WebUI in the foreground with ironclaw serve. Slack and Telegram require extension installation and setup on the WebUI Extensions page. Configuration writes do not automatically restart the service.

Who IronClaw is for

IronClaw is intended for people seeking a personal AI assistant with local application storage and configurable tool access. The documented extension system also supports developers who want to add tools or channels. Background routines, workspace files, and persistent preferences support recurring personal and professional tasks rather than only isolated chat exchanges.

  • IronClaw supports scheduled monitoring and maintenance through proactive background execution.
  • IronClaw handles automation triggered by incoming events or webhooks.
  • IronClaw stores notes, logs, and context in a path-based workspace.
  • IronClaw maintains personality and preferences across sessions using identity files.
  • IronClaw supports assistant access through terminal, browser, and messaging channels.

IronClaw licensing and access

IronClaw is distributed as an open-source project under a choice of the Apache License or MIT License. The repository provides pre-built binaries, release installers, and source-build instructions rather than a published subscription plan. Documented installation paths cover macOS, Linux, Windows, and Windows/WSL, with guided onboarding used for initial configuration.

IronClaw strengths and limitations

IronClaw combines local application storage, encrypted secrets, and an audit log of tool executions with several execution safeguards. WASM permissions, endpoint allowlisting, request and response leak scans, and resource limits constrain tool activity. The source also states that IronClaw has no telemetry or analytics. These controls are described as defense in depth, not a guarantee against every security risk.

  • IronClaw onboarding requires selecting an LLM provider and entering an API key.
  • IronClaw source builds require Rust and Node.js with Corepack/pnpm.
  • IronClaw requires a service restart after configuration changes that affect the running service.
  • IronClaw messaging channels require separate extension setup in the WebUI.

IronClaw FAQ

What is IronClaw?

IronClaw is an open-source personal AI assistant and Agent OS focused on privacy, security, and extensibility. Written in Rust and inspired by OpenClaw, IronClaw combines sandboxed tools, background automation, persistent memory, and browser or terminal access.

How is IronClaw licensed and accessed?

IronClaw is available as an open-source project under a choice of the Apache License or MIT License. IronClaw provides pre-built binaries, installers, and source-build instructions through its repository. Guided setup requires selecting an LLM provider and entering its API key.

What security features does IronClaw provide?

IronClaw provides WASM sandboxing, capability-based permissions, endpoint allowlisting, encrypted credential storage, and leak detection. IronClaw also applies prompt injection detection, content sanitization, and policy rules to external content, and maintains an audit log of tool executions.

Does IronClaw support MCP, Slack, and Telegram?

IronClaw connects to Model Context Protocol servers and supports messaging channels including Slack and Telegram through extensions. IronClaw requires users to install those channel extensions and complete setup on the WebUI Extensions page rather than enabling them through configuration-file settings.

How do you install IronClaw?

IronClaw can be installed from a selected repository release using pre-built binaries or installers, or built from source. IronClaw documents macOS, Linux, and Windows installation paths. The ironclaw onboard command then configures an LLM provider, encrypted credentials, and WebUI access.

From the community

No reviews yet

Be the first to review this tool. What you write helps the next person choose.

More tools like IronClaw

nanobot

www.github.com

Self-hosted personal AI agent for file tasks, persistent memory, scheduled automation, and chat apps.

Free

Hermes Agent

hermes-agent.nousresearch.com

Open-source AI agent with persistent memory, scheduled tasks, messaging integrations, and sandboxed execution.

Free

CodeWhale

www.github.com

Open-source terminal coding agent that reads repositories, edits files, and runs commands with cloud or local models.

Free

Background Agents

www.github.com

Open-source background coding agents with shared sessions, development sandboxes, and scheduled automations.

Free