ICIronClaw logo
Chatbots

IronClaw

Open-source AI agent for busywork running in encrypted enclaves with a credential vault.

At a glance

Free + paid · Source review 2026-11-04
Price & billing

Starter $0 per month verified: 1 agent instance in a secure environment, $5 credits included. Paid tiers are page observations in $ without a named currency, so no paid-tier charge fact is recorded: Basic $20 per month with up to 2 agent instances and $20 credits included, Pro Plus $200 per month with up to 5 agent instances and $200 credits included.

Free allowance

Not separately confirmed. Check the current plan before relying on a free allowance.

Best suited to

Users running OpenClaw with sensitive credentials who want prompt injection and malicious-skill exfiltration blocked architecturally.; People who want an agent to triage email, brief them each morning, prepare for meetings, and monitor endpoints, repos, and keywords from chat.; Teams that want model-agnostic agents, with compatibility listed for Anthropic, OpenAI, GitHub Copilot, Google Gemini, MiniMax, Mistral, Ollama, OpenRouter, Together AI, and Fireworks AI.

Fit is editorial guidance, not a measured result.
Platforms

Access: not separately verified. Operating systems not separately verified.

Published-source facts and editorial guidance. No product test is implied. Confirm current conditions before purchase.

No product screenshot is claimed for this listing. Use the official website ↗ to inspect the current interface.

What to know before choosing

Good fit

Users running OpenClaw with sensitive credentials who want prompt injection and malicious-skill exfiltration blocked architecturally.; People who want an agent to triage email, brief them each morning, prepare for meetings, and monitor endpoints, repos, and keywords from chat.; Teams that want model-agnostic agents, with compatibility listed for Anthropic, OpenAI, GitHub Copilot, Google Gemini, MiniMax, Mistral, Ollama, OpenRouter, Together AI, and Fireworks AI.

Limits and trade-offs

Tools that need filesystem access or outbound connections beyond an allowlist, since every tool runs in its own WebAssembly sandbox with no filesystem access and no outbound connections beyond the allowlist.; Users who need more than 1 agent instance on Starter or 2 on Basic, since Starter activates 1 agent instance and Basic allows up to 2, with up to 5 on Pro Plus.; Users who need credentials visible to the model itself, since credentials are invisible to the AI and are injected only for endpoints approved in the allowlist.

Plans and free allowances

Starter $0 per month verified: 1 agent instance in a secure environment, $5 credits included. Paid tiers are page observations in $ without a named currency, so no paid-tier charge fact is recorded: Basic $20 per month with up to 2 agent instances and $20 credits included, Pro Plus $200 per month with up to 5 agent instances and $200 credits included.

Evidence

Official homepage ↗ (reviewed 2026-11-04)

Official published documentation reviewed; not a product test

Dates: Last reviewed 2026-11-04. Last reviewed is the latest source or editorial review date; it is not a content-change date.

A closer look at IronClaw

IronClaw is an open-source AI agent for busywork that runs in encrypted enclaves on NEAR AI Cloud or locally, so an OpenClaw assistant can automate email, calendar, chat, and monitoring work while credentials stay in an encrypted vault.

Editorial guidance based on published documentation. The workflow below is a suggested starting point, not a measured product result.

When to choose IronClaw

  • Users running OpenClaw with sensitive credentials who want prompt injection and malicious-skill exfiltration blocked architecturally.
  • People who want an agent to triage email, brief them each morning, prepare for meetings, and monitor endpoints, repos, and keywords from chat.
  • Teams that want model-agnostic agents, with compatibility listed for Anthropic, OpenAI, GitHub Copilot, Google Gemini, MiniMax, Mistral, Ollama, OpenRouter, Together AI, and Fireworks AI.

When to choose something else

  • Tools that need filesystem access or outbound connections beyond an allowlist, since every tool runs in its own WebAssembly sandbox with no filesystem access and no outbound connections beyond the allowlist.
  • Users who need more than 1 agent instance on Starter or 2 on Basic, since Starter activates 1 agent instance and Basic allows up to 2, with up to 5 on Pro Plus.
  • Users who need credentials visible to the model itself, since credentials are invisible to the AI and are injected only for endpoints approved in the allowlist.

A practical starting workflow

  1. Deploy an IronClaw instance in one click on NEAR AI Cloud, where it boots inside a Trusted Execution Environment.
  2. Store API keys, tokens, and passwords in the encrypted vault, with injection only for endpoints you have approved.
  3. Hand off busywork in chat, such as inbox triage, morning briefings, meeting prep, and monitors, and let the agent run on its own.

Check before committing

  • Plan amounts on the homepage pricing section are shown in $ without a named currency, so no paid-tier charge fact is recorded; only the Starter $0 per month plan is recorded as a structured fact.
  • The guessed pricing and FAQ paths return 404 pages, so pricing facts are drawn from the homepage pricing section, not a separate pricing page.

Common questions

What is IronClaw?

IronClaw is an open-source agent for busywork, in encrypted enclaves, where secrets never touch the model, built for the OpenClaw personal AI assistant and powered by NEAR AI Cloud or run locally.

Official homepage ↗

How does IronClaw protect credentials?

Credentials are invisible to the AI. API keys, tokens, and passwords are encrypted at rest and injected into requests at the host boundary, only for endpoints you have approved.

Official homepage ↗

How do I deploy IronClaw?

Launch an IronClaw instance on NEAR AI Cloud, where it boots inside a Trusted Execution Environment, encrypted from the start with no setup required, then add credentials to the encrypted vault.

Official homepage ↗

Compare approaches for this task

  • AIOProductOS — Product teams that want to build what pays, since the platform ties each customer record to feedback, revenue, work and code.; Companies that need enterprise controls, since the pricing page highlights EU and US data residency, BYOK encryption, SAML/SSO on Pro and 150+ compliance controls.
  • AISOAP — Clinicians who spend too long on documentation, since AISOAP records, transcribes and generates SOAP notes from the visit conversation.; Clinics that need team controls, since multi-person plans add team management, template creation and an organisation-wide BAA.

Sources and checks

Last source review is not a date of product testing. No rating or test result is implied by inclusion. Check the current plan and terms before buying.