The deal between us v2.0

Last updated: March 26, 2026 · Effective: April 26, 2026 · Replaces v1.0 (March 26, 2026)

The short version 🦖: I'm shipped as self-hosted software now — you run the instance, you control the data, and you're the data controller for your deployment. My outputs still need a human glance before anything important. The legal version starts below.

1. Acceptance

By accessing or using TaskZilla, you agree to be bound by these Terms of Service, our Privacy Policy, and our Security Policy. If you do not agree, do not use the software. If you are accepting on behalf of an organization, you represent and warrant that you have authority to bind that organization.

2. The Product

TaskZilla is a self-hosted AI-powered project management software product distributed as a deployable blueprint. It automates standups, routes tasks, manages team memory, and integrates with your existing tools (ClickUp, Telegram, GitHub, and others).

2.1 Deployment model

TaskZilla is self-hosted. The deploying organization installs, configures, and operates their own instance on their own infrastructure. TaskZilla B.V. does not host customer instances, does not have access to customer data, and does not operate as a data processor for deployed instances.

2.2 Software delivery

We provide the software on an "as-is" and "as-available" basis and continuously improve it. Updates are distributed as versioned releases. The deploying organization decides when and whether to apply updates.

3. Artificial Intelligence Disclosure

3.1 AI system

TaskZilla uses large language models and machine learning systems to generate content (including standup reports, task summaries, and recommendations), route tasks, assign priorities, and process team communications. All outputs generated by TaskZilla are produced by artificial intelligence and should be reviewed by a human before being relied upon for consequential decisions. TaskZilla does not make employment, performance evaluation, or human resources decisions.

3.2 AI risk classification

TaskZilla is classified as a limited-risk AI system under Regulation (EU) 2024/1523 (the EU Artificial Intelligence Act). TaskZilla is not designed or intended for use as a high-risk AI system. Specifically, TaskZilla does not:

• Make or inform employment, recruitment, or termination decisions.
• Evaluate individual worker performance for HR purposes.
• Allocate tasks based on profiling of individual behavior or personal traits.
• Perform biometric identification or emotion recognition.
• Operate in any domain listed in Annex III of the EU AI Act.

3.3 AI output accuracy

AI-generated content may contain errors, inaccuracies, hallucinations, or omissions. TaskZilla's outputs are provided for informational and productivity purposes only. You are responsible for verifying AI-generated content before acting upon it. We do not guarantee the accuracy, completeness, or fitness for any particular purpose of AI-generated outputs.

3.4 Human oversight and HITL controls

All automated actions taken by TaskZilla are subject to human oversight:

• Workspace administrators may configure, override, or disable any automated function.
• Task assignments and priority changes made by TaskZilla can be modified or reversed by any authorized user.
• AI-generated content is presented for human review before distribution where configured.
• No automated decision is irrevocable — users retain full control over their workspace and tasks at all times.

Additionally, TaskZilla enforces a Human-in-the-Loop (HITL) gate for high-risk actions. Every action is scored against three dimensions — reversibility, blast radius, and data sensitivity — on a 0–6 scale. Actions scoring 4 or above (including data deletion, secrets modification, external message delivery, bulk operations, credential changes, and server-side script execution) require explicit human approval before execution. If no approval is received within 5 minutes (configurable by workspace administrators), the action is denied by default. Medium-risk actions (score 2–3) are configurable per workspace. All actions and HITL decisions are logged in an immutable audit trail. See our Privacy Policy Section 5.3 for full details.

3.5 Not a substitute for human judgment

TaskZilla's outputs and recommendations should never be the sole basis for consequential decisions affecting individuals, including employment, termination, compensation, performance evaluation, or legal determinations. Always apply human judgment. TaskZilla is a productivity tool — not a decision-maker.

3.6 No training on your data

We do not use your workspace content, prompts, or outputs to train, fine-tune, or improve any AI models. Your data is processed solely to deliver the service to you.

4. Accounts

• You must provide accurate information during onboarding.
• You are responsible for maintaining the security of your account credentials and API tokens.
• One person or bot per account — shared accounts are not permitted.
• You must promptly notify us of any unauthorized access to your account.
• You are responsible for all activity that occurs under your account.

5. Acceptable Use

You agree not to:

• Use TaskZilla for unlawful purposes or to violate others' rights.
• Attempt to reverse-engineer, probe, or exploit the platform's infrastructure.
• Exceed reasonable usage limits or use automated tools to abuse the API.
• Transmit malware, spam, or harmful content through integrations.
• Use the service to process special category data (Art. 9 GDPR) unless expressly agreed in a separate Data Processing Agreement.
• Use TaskZilla's AI capabilities to generate content that is deceptive, discriminatory, defamatory, or unlawful.
• Resell, sublicense, or redistribute the service without prior written consent.

6. Prohibited AI Uses and EU AI Act High-Risk Safeguards

6.1 Prohibited practices (Article 5, EU AI Act)

You shall not use TaskZilla for any purpose classified as an unacceptable risk or prohibited practice under Article 5 of the EU AI Act, including but not limited to:

• Social scoring or behavioral credit systems.
• Real-time or remote biometric identification in public spaces.
• Subliminal, manipulative, or deceptive techniques that cause harm.
• Exploitation of vulnerabilities of specific groups (age, disability, social situation).

These uses are blocked outright — no HITL override is possible.

6.2 High-risk use cases (Annex III, EU AI Act) — mandatory HITL gate

Any user-driven action that falls within or reasonably could be interpreted as a high-risk AI use case as defined in Annex III of the EU AI Act is automatically routed through the HITL gate and requires explicit human approval before execution. This includes but is not limited to actions that could:

• Make or substantially inform employment, recruitment, or termination decisions — including task allocation patterns that could be interpreted as performance evaluation or workforce management.
• Produce legal effects or similarly significant effects on individuals — including automated decisions affecting access, eligibility, or resource allocation.
• Profile, score, or rank individuals based on behavior, performance, or personal characteristics — including workload-based team member comparisons or productivity metrics.
• Affect access to education, vocational training, or certification — including decisions about learning resources, course access, or training eligibility.
• Affect access to essential services — including access to tools, platforms, or information that individuals depend on for their work or livelihood.
• Inform decisions about creditworthiness, insurance, or eligibility determinations of any kind.
• Support or inform judicial, legal, or administrative proceedings — including evidence gathering, case preparation, or compliance determinations.

TaskZilla's HITL system detects these patterns through action classification. When an action's context matches a high-risk indicator, it is escalated regardless of its base risk score — meaning even a normally low-risk action (e.g., a task query) is gated if it is being used in a high-risk context (e.g., generating a report that ranks team members by output). The HITL gate cannot be bypassed or disabled for EU AI Act high-risk use cases.

6.3 Other prohibited uses

You shall not use TaskZilla:

• For any purpose classified as high-risk under the EU AI Act without prior written agreement and, where applicable, appropriate conformity assessment — even with HITL approval.
• To generate deepfakes, impersonate individuals, or create misleading content that could be mistaken for human-produced work in contexts where disclosure is required.
• In any manner that violates applicable law, including the EU AI Act, the GDPR, or national implementing legislation.

7. Intellectual Property

7.1 Your content

You retain ownership of all content you create, upload, or import through TaskZilla. By using the service, you grant us a limited, non-exclusive, revocable license to process your content solely to provide and operate the service. This license terminates when your account is closed.

7.2 AI-generated outputs

Outputs generated by TaskZilla's AI (standup reports, summaries, recommendations) are derived from your content and third-party AI models. You may use these outputs for your business purposes. We make no claim of ownership over AI-generated outputs, but we also make no representation that such outputs are protectable by intellectual property rights or that they do not infringe third-party rights.

7.3 Our platform

We retain ownership of the TaskZilla platform, its code, models, design, documentation, and branding. Nothing in these terms grants you any rights to our intellectual property beyond the limited license to use the service.

7.4 Feedback

If you voluntarily provide feedback, suggestions, or ideas, we may use them to improve the service without obligation or compensation to you.

8. Billing and Cancellation

• Free-tier users may use the platform within published limits at no charge.
• Paid plans are billed monthly or annually via Stripe. Prices are listed on our Pricing page and are in EUR unless stated otherwise.
• You may cancel at any time. Access continues until the end of your current billing period. No partial refunds for unused time.
• We may change pricing with at least 30 days' notice. Price changes do not apply retroactively to the current billing period. If you do not accept a price change, you may cancel before it takes effect.
• Refunds are handled on a case-by-case basis — contact support.
• If payment fails, we will attempt to charge your payment method for up to 14 days. If payment remains unsuccessful, we may downgrade your account to the free tier.

9. Data and Privacy

Your use of the TaskZilla website is governed by our Privacy Policy, which describes how we collect and use website visitor data and describes the data processing capabilities built into the software.

For deployed instances: The deploying organization is the data controller and is solely responsible for:

• Establishing the legal basis for processing personal data within their instance.
• Signing Data Processing Agreements with their chosen third-party providers (LLM APIs, cloud hosting, project management tools, etc.).
• Ensuring adequate transfer mechanisms (e.g., SCCs) for any international data transfers.
• Fulfilling data subject rights requests from their users.
• Publishing their own privacy policy covering their TaskZilla deployment.

TaskZilla B.V. does not process, access, or store any data from customer deployments.

10. Confidentiality

TaskZilla B.V. does not have access to your deployed instance or workspace data. Your data remains entirely within your infrastructure. We do not use any customer content to train AI models. Where you share data with us through support requests, we treat it as confidential.

11. Third-Party Integrations

TaskZilla integrates with third-party services (ClickUp, Telegram, GitHub, LLM APIs, etc.) chosen and configured by the deploying organization. Your use of these integrations is subject to the respective third-party terms and privacy policies. TaskZilla B.V. is not responsible for the availability, accuracy, or practices of third-party services. You are responsible for configuring integrations appropriately, for the API tokens you connect, and for signing any required Data Processing Agreements with those providers.

12. Availability and SLA

We aim for high availability but do not guarantee uninterrupted service. Planned maintenance will be communicated in advance where possible. See our Status page for real-time uptime information. Specific SLA terms may be agreed separately for enterprise customers.

13. Warranties and Disclaimers

THE SERVICE IS PROVIDED "AS IS" AND "AS AVAILABLE." TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, WE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT.

Without limiting the foregoing:

• We do not warrant that AI-generated outputs will be accurate, complete, free from bias, or fit for any purpose.
• We do not warrant uninterrupted or error-free operation of the service.
• We do not warrant that the service will meet your specific requirements.
• We do not warrant that third-party integrations will function correctly or remain available.

Nothing in these terms excludes or limits liability that cannot be excluded or limited under applicable law, including liability for fraud, gross negligence, or death or personal injury caused by our negligence.

14. Limitation of Liability

To the maximum extent permitted by applicable law:

• TaskZilla's total aggregate liability for any and all claims arising out of or relating to these terms or the service is limited to the amount you paid us in the 12 months preceding the claim, or EUR 100, whichever is greater.
• We are not liable for indirect, incidental, special, consequential, or punitive damages, including but not limited to loss of profits, data, business, or goodwill, regardless of the theory of liability.
• We are not liable for any damages arising from: (a) your reliance on AI-generated outputs; (b) your failure to verify AI outputs before acting on them; (c) your use of the service for any high-risk or prohibited purpose; (d) third-party service failures or integration issues.

15. Indemnification

You agree to indemnify and hold harmless TaskZilla B.V., its officers, directors, and employees from any claim, demand, or damage (including reasonable legal fees) arising from: (a) your use of the service in violation of these terms; (b) your use of AI-generated outputs; (c) your use of the service for any prohibited purpose under Section 6; (d) your violation of applicable law; (e) any claim by a third party resulting from content you created or processed through the service.

16. Termination

• You may delete your account at any time via support.
• We may suspend or terminate your account if you materially breach these terms, with reasonable notice where practicable (except in cases of abuse or security threats, where immediate suspension may be necessary).
• Upon termination, your right to use the service ceases. You may request an export of your data within 30 days of termination. We will retain your data only as described in our Privacy Policy retention schedule.
• Sections 7, 13, 14, 15, and 18 survive termination.

17. Force Majeure

Neither party is liable for failure to perform obligations due to circumstances beyond its reasonable control, including natural disasters, pandemics, war, terrorism, government actions, power failures, internet disruptions, or third-party service outages. The affected party must notify the other promptly and use reasonable efforts to mitigate the impact.

18. Changes

We may update these terms. Material changes will be communicated at least 30 days in advance via email or in-app notification. Continued use after changes take effect constitutes acceptance. If you do not agree to updated terms, you may terminate your account before the changes take effect. Previous versions are archived and available upon request.

19. Governing Law and Disputes

These terms are governed by the laws of the Netherlands, without regard to conflict of law principles. Any disputes arising from these terms or the service will be resolved exclusively in the courts of Amsterdam, the Netherlands.

If you are a consumer in the EU, you also have the right to submit complaints to the European Commission's Online Dispute Resolution platform at ec.europa.eu/consumers/odr. Nothing in these terms affects your rights as a consumer under mandatory consumer protection legislation in your country of residence.

20. EU AI Act Compliance

TaskZilla is provided in compliance with Regulation (EU) 2024/1523 (the EU Artificial Intelligence Act). As a limited-risk AI system, TaskZilla meets all applicable transparency obligations under Article 50, including:

• Clear disclosure that users are interacting with an AI system.
• Marking of AI-generated content as artificially generated.
• Informing users of the existence and logic of automated processing.

TaskZilla is not classified as a high-risk AI system and is not used for any prohibited AI practice under Article 5 of the EU AI Act. Users must not deploy TaskZilla for any high-risk purpose as defined in Annex III without prior written agreement.

21. Severability

If any provision of these terms is held to be invalid or unenforceable, the remaining provisions continue in full force and effect. The invalid provision will be modified to the minimum extent necessary to make it valid and enforceable.

22. Entire Agreement

These terms, together with the Privacy Policy, Security Policy, and any separate agreements (e.g., enterprise SLAs, data processing agreements), constitute the entire agreement between you and TaskZilla B.V. regarding the service. Any prior agreements or understandings are superseded.

Still have questions? Ping me at support.taskzilla.ai — I read every one.