MindClaw Terms of Service

Please read these terms carefully before using MindClaw.

Last updated: March 2026

1. Acceptance of Terms

By installing, accessing, or using MindClaw ("the Software"), you agree to be bound by these Terms of Service ("Terms"). MindClaw is a product of Liberation Softwares LLP ("we", "us", "our"), the makers of Elephas. If you do not agree to these Terms, do not install or use the Software.

2. Description of Service

MindClaw is a native macOS application that runs AI-powered agents locally on your computer. These agents can perform automated tasks including but not limited to: accessing your calendar and reminders, reading and organizing files, monitoring websites, executing shell commands (with your approval), browsing the web, and interacting with third-party APIs.

The Software requires you to provide your own API keys for third-party AI providers (such as OpenAI, Anthropic, or others). We do not provide AI inference services and are not responsible for the availability, pricing, or terms of these third-party providers.

3. AI Agent Behavior Disclaimer

Important: Please read this section carefully.

MindClaw uses artificial intelligence models provided by third parties to power its agents. AI agents can and will occasionally produce incorrect, unexpected, or unintended results. While we implement multiple layers of safety controls (approval gates, command safety filters, risk-level assessments), no safety system is perfect.

By using MindClaw, you acknowledge and agree that:

  • AI agents may take actions that you did not intend or expect, even in autonomous mode
  • Agents may misinterpret instructions, produce inaccurate information, or execute tasks incorrectly
  • Shell commands executed by agents, even after passing safety checks, could modify or delete files or data on your system
  • Scheduled (heartbeat) agents run automatically and may take actions while you are not actively monitoring
  • You are solely responsible for reviewing, approving, and supervising agent actions, especially when operating in autonomous mode
  • You should maintain regular backups of any data that agents can access

4. Limitation of Liability

TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, LIBERATION SOFTWARES LLP SHALL NOT BE LIABLE FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, OR PUNITIVE DAMAGES, INCLUDING BUT NOT LIMITED TO:

  • Loss of data, including files deleted, modified, moved, or corrupted by AI agent actions
  • Loss of revenue or business resulting from incorrect agent actions or inaccurate information
  • System damage resulting from shell commands, scripts, or other operations executed by agents
  • Missed events, deadlines, or reminders due to agent scheduling failures, system sleep/wake issues, or incorrect calendar/reminder interactions
  • Unauthorized or unintended actions taken by agents operating in autonomous mode
  • Costs incurred from third-party AI API usage, including unexpected token consumption
  • Privacy breaches resulting from data sent to third-party AI providers through agent operations

IN NO EVENT SHALL OUR TOTAL LIABILITY EXCEED THE AMOUNT YOU PAID FOR THE SOFTWARE IN THE TWELVE (12) MONTHS PRECEDING THE CLAIM.

5. Disclaimer of Warranties

THE SOFTWARE IS PROVIDED "AS IS" AND "AS AVAILABLE" WITHOUT WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT.

We do not warrant that: (a) the Software will be error-free or uninterrupted; (b) AI agents will always behave as intended; (c) safety filters will catch all dangerous commands; (d) scheduled tasks will execute at the exact specified time; (e) data processed by agents will be accurate or complete; or (f) the Software will be compatible with all system configurations.

6. Your Responsibilities

As a user of MindClaw, you agree to:

  • Maintain backups of all important data on your system before enabling agent access to files, calendar, reminders, or shell commands
  • Review agent actions regularly, especially when using autonomous execution mode
  • Configure appropriate execution policies — use approval-required mode for high-risk actions
  • Secure your API keys and not share them with unauthorized parties
  • Comply with the terms of service of any third-party AI providers whose APIs you use through MindClaw
  • Not use MindClaw for any illegal, harmful, or malicious purpose
  • Not circumvent or attempt to bypass the Software's safety mechanisms (CommandSafetyGate, ApprovalGate, or risk-level filters)
  • Monitor token usage and understand that AI API calls incur costs from your chosen provider

7. Third-Party Services

MindClaw integrates with third-party services that are governed by their own terms:

  • AI Providers (OpenAI, Anthropic, OpenRouter, Ollama) — API calls are made directly from your Mac to these providers using your API keys. Their terms of service and privacy policies apply to all data sent through their APIs.
  • Telegram (optional) — If you enable Telegram relay, messages are routed through a Cloudflare Worker relay service operated by us. See our Privacy Policy for details.
  • Dodo Payments — License purchases and validation are processed through Dodo Payments.

We are not responsible for the availability, accuracy, or security of third-party services.

8. License Grant

Subject to these Terms, we grant you a limited, non-exclusive, non-transferable, revocable license to use MindClaw on macOS devices that you own or control. This license does not include the right to:

  • Modify, reverse engineer, decompile, or disassemble the Software
  • Redistribute, sublicense, or resell the Software
  • Remove or alter any proprietary notices or labels
  • Use the Software to build a competing product

9. Termination

We may suspend or terminate your license at any time if you violate these Terms. Upon termination, you must cease all use of the Software and delete all copies. Your agent data, memory files, and conversations stored locally on your Mac remain yours — we cannot access or delete them.

10. Changes to These Terms

We may update these Terms from time to time. We will notify you of significant changes through the Software or via email. Continued use of MindClaw after changes constitutes acceptance of the updated Terms.

11. Governing Law

These Terms are governed by the laws of India. Any disputes shall be resolved in the courts of Bangalore, Karnataka, India.

12. Contact Us

If you have questions about these Terms, contact us at support@elephas.app.

Liberation Softwares LLP
Bangalore, Karnataka, India

← Back to MindClaw | Privacy Policy →