← Back to Agent Workspace

Agent Workspace

Agent Workspace Terms of Use

Effective date: March 18, 2026

These Terms of Use govern your access to and use of the Agent Workspace website, downloads, packaged desktop application, release artifacts, update feeds, documentation, and related materials that the Agent Workspace maintainers make available.

By accessing, downloading, installing, or using Agent Workspace, you agree to these Terms. If you do not agree, do not use Agent Workspace.

1. Who we are

Agent Workspace is published by the Agent Workspace maintainers. In these Terms, "Agent Workspace," "we," "our," and "us" refer to the publisher and maintainers of Agent Workspace.

Current public contact channels:

2. What Agent Workspace is

Agent Workspace is local-first developer tooling. It can orchestrate terminal-based agents, repositories, worktrees, local services, command execution, prompts, reviews, browser-like workspace tabs, dashboards, and optional service-backed features such as GitHub-linked repository and pull-request workflows, Trello task workflows, optional speech-recognition backends, and provider-backed AI helpers.

Because Agent Workspace can launch commands, read and write files, and trigger actions through connected tools, it is powerful software that can materially affect your machine, repositories, services, secrets, and data.

3. Eligibility and authority

You may use Agent Workspace only if:

  • you are legally capable of entering into these Terms;
  • you are authorized to use the machines, repositories, credentials, and data you connect;
  • your use complies with all applicable laws, regulations, contracts, and workplace policies.

If you use Agent Workspace for an organization, you represent that you have authority to bind that organization to these Terms.

4. Relationship to open-source licenses

The Agent Workspace repository source code is made available under the MIT License, and the application may include third-party open-source components under their own licenses.

These Terms do not revoke, narrow, or override any rights granted to you under the applicable open-source licenses for the source code or third-party components. If these Terms conflict with an applicable open-source license for a particular component, the open-source license controls for that component.

These Terms do apply to:

  • the packaged desktop application and release materials as distributed by us;
  • the Agent Workspace website and branding;
  • any hosted update or distribution channels we operate;
  • our disclaimers, liability terms, usage rules, and product-specific conditions to the extent permitted by law.

5. License to use Agent Workspace

Subject to these Terms, we grant you a limited, non-exclusive, revocable license to access and use Agent Workspace.

You may:

  • download, install, and run the packaged app;
  • run the software from source;
  • use the website and documentation;
  • use, modify, and redistribute source code to the extent permitted by the MIT License and other applicable open-source licenses.

6. Acceptable use

You may not use Agent Workspace to:

  • access systems, repositories, or data without authorization;
  • violate law, sanctions, export controls, intellectual-property rights, privacy rights, employment obligations, or contractual restrictions;
  • distribute malware, destructive code, credential theft tooling, or deceptive content;
  • interfere with, disable, or bypass security controls unless you are expressly authorized to do so;
  • misrepresent AI-generated output as human-reviewed, security-audited, or professionally verified when it is not;
  • operate in a manner that is abusive, fraudulent, or harmful to others.

7. High-risk and prohibited scenarios

Agent Workspace is not designed, tested, or licensed for life-critical or safety-critical use. You must not use it in contexts where failure, hallucination, bad automation, or command execution could reasonably lead to death, bodily injury, environmental damage, or major property loss, including:

  • medical diagnosis or treatment;
  • emergency response;
  • aviation, autonomous vehicle, or transportation safety systems;
  • nuclear, weapons, or critical infrastructure control systems;
  • legally binding professional advice without qualified human review.

You remain solely responsible for human review, approvals, testing, staging, backups, and rollback procedures before using output generated or actions initiated through Agent Workspace.

8. Your responsibilities

You are solely responsible for:

  • every prompt, command, script, automation, workflow, or policy you run through Agent Workspace;
  • reviewing AI-generated text, code, diffs, comments, task updates, and automation outcomes before relying on them;
  • securing your machine, local network, repositories, browser session, API keys, access tokens, environment variables, and credentials;
  • configuring safe permissions, backups, branch protections, access controls, and deployment safeguards;
  • using only trusted repositories, workspaces, and folders;
  • ensuring that your use is lawful and authorized.

9. Third-party tools and service-backed features

Agent Workspace can work with third-party tools and services, including but not limited to GitHub, Trello, Anthropic, OpenAI, Google, local CLIs, and other agent providers.

Examples of optional service-backed features in the current product include GitHub repository discovery and pull-request workflows, Trello board and card workflows, optional browser speech recognition, and provider-backed AI requests.

Those third parties are independent from us. They have separate terms, privacy policies, availability, quotas, pricing, and security practices. We are not responsible for:

  • third-party outages, errors, suspensions, or pricing changes;
  • model output or provider moderation decisions;
  • loss, disclosure, corruption, or misuse of data caused by third-party services;
  • actions you authorize Agent Workspace to take with those providers.

10. AI-specific disclosures

Agent Workspace can assist with prompts, automations, reviews, code generation, summaries, classifications, and action routing, but it does not guarantee correctness, fitness, legal compliance, security, or suitability for any purpose.

AI output may be incomplete, biased, insecure, outdated, or wrong. You must independently review and validate all output before relying on it, shipping it, publishing it, or using it to make operational, legal, security, financial, or compliance decisions.

Agent Workspace and its outputs are not legal, medical, tax, accounting, security, or other professional advice.

11. Website, branding, and trademarks

The Agent Workspace name, logos, website design, branding, and related trade dress may be protected by trademark, copyright, unfair competition, and other laws to the extent applicable.

These Terms do not grant you any right to use our trademarks, logos, or branding in a way that suggests endorsement, affiliation, sponsorship, or official status, except as otherwise permitted by law or by separate written permission.

12. Local-first operation and privacy

Agent Workspace is designed to run on your own hardware and does not require publisher-hosted user accounts or publisher-hosted telemetry by default.

That said, the product still processes and stores data locally on your device, and it can transmit data to third-party services when you configure or use optional features that depend on those services. Our Privacy Policy explains this in more detail:

Agent Workspace Privacy Policy

13. Releases, verification, and availability

We may change, patch, improve, suspend, or discontinue any part of Agent Workspace or its distribution channels at any time, with or without notice.

We are not obligated to provide maintenance, support, updates, or any minimum service level. Features may change between source builds, local runs, preview builds, and tagged releases.

Before running packaged releases, you should verify the release asset you downloaded against the published release information. That includes checking the published SHA-256 digest and, if a platform signature is present, checking that signature as well. If verification fails, do not run the file.

14. Feedback

If you submit suggestions, ideas, bug reports, feature requests, or other feedback, you grant us a worldwide, perpetual, irrevocable, sublicensable, royalty-free license to use, modify, publish, and incorporate that feedback without restriction or compensation, unless applicable law prohibits that result.

15. Disclaimer of warranties

TO THE MAXIMUM EXTENT PERMITTED BY LAW, AGENT WORKSPACE IS PROVIDED "AS IS" AND "AS AVAILABLE," WITHOUT WARRANTIES OF ANY KIND, WHETHER EXPRESS, IMPLIED, STATUTORY, OR OTHERWISE, INCLUDING ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE, NON-INFRINGEMENT, ACCURACY, SECURITY, RELIABILITY, QUIET ENJOYMENT, OR THAT THE SOFTWARE OR WEBSITE WILL BE UNINTERRUPTED, ERROR-FREE, OR FREE OF HARMFUL COMPONENTS.

16. Non-excludable rights

Nothing in these Terms excludes, restricts, or modifies any rights or remedies that cannot be excluded, restricted, or modified under applicable law.

17. Limitation of liability

TO THE MAXIMUM EXTENT PERMITTED BY LAW, THE AGENT WORKSPACE MAINTAINERS AND THEIR LICENSORS, CONTRIBUTORS, CONTRACTORS, AND SUPPLIERS WILL NOT BE LIABLE FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, EXEMPLARY, OR PUNITIVE DAMAGES, OR FOR ANY LOSS OF PROFITS, REVENUE, BUSINESS, GOODWILL, DATA, SECRETS, REPOSITORIES, OR SYSTEM AVAILABILITY, ARISING OUT OF OR RELATING TO AGENT WORKSPACE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

TO THE MAXIMUM EXTENT PERMITTED BY LAW, THE TOTAL AGGREGATE LIABILITY OF THE AGENT WORKSPACE MAINTAINERS AND THEIR LICENSORS, CONTRIBUTORS, CONTRACTORS, AND SUPPLIERS FOR ALL CLAIMS ARISING OUT OF OR RELATING TO AGENT WORKSPACE WILL NOT EXCEED THE GREATER OF:

  • the amount you paid us for the applicable product in the twelve months before the event giving rise to the claim; or
  • U.S. $50.

Some jurisdictions do not allow certain limitations, so parts of this section may not apply to you.

18. Indemnification

To the maximum extent permitted by law, you will defend, indemnify, and hold harmless the Agent Workspace maintainers and their licensors, contributors, contractors, and suppliers from claims, damages, losses, liabilities, judgments, penalties, and expenses, including reasonable attorneys' fees, arising out of or related to:

  • your misuse of Agent Workspace;
  • your violation of these Terms;
  • your violation of law or third-party rights;
  • your unauthorized access to, or use of, systems, repositories, credentials, or data;
  • your content, prompts, automations, scripts, or workflows.

19. Suspension and termination

We may suspend or terminate access to websites, downloads, update feeds, or other materials we control if we reasonably believe your use creates legal, security, abuse, or operational risk.

You may stop using Agent Workspace at any time. Sections that by their nature should survive termination will survive, including sections concerning open-source licensing, disclaimers, limitations of liability, indemnity, feedback, and general legal interpretation.

20. Export and sanctions compliance

You may not use, export, re-export, transfer, or make Agent Workspace available in violation of applicable export control, sanctions, or trade laws.

21. Changes to these Terms

We may update these Terms from time to time. When we do, we will update the effective date above. If you continue to use Agent Workspace after updated Terms take effect, you agree to the updated Terms to the extent permitted by law.

22. General

If any provision of these Terms is held unenforceable, the remaining provisions will remain in effect to the maximum extent permitted by law.

Our failure to enforce a provision is not a waiver of that provision.

23. Contact

Questions about these Terms can be directed through: