Acceptable Use Policy
Last updated: Nov 15, 2026
This Acceptable Use Policy (“AUP”) defines acceptable and prohibited behavior on the agenticml-lite hosted service (https://app.agenticml.xyz), operated by the agenticML project (“agenticML”, “we”, “us”, or “our”).
This policy applies to all users, accounts, API keys, and machine sync traffic. By using the service, you agree to comply with this policy and our Terms of Service.
1. Intended Use
Section titled “1. Intended Use”The service is provided exclusively for machine learning experiment tracking, logging metrics, recording configurations, tracking code lineage, and storing artifacts using the open-source agenticml-lite library.
2. Prohibited Activities
Section titled “2. Prohibited Activities”You may not use the service to:
Security violations and attacks
Section titled “Security violations and attacks”- Probe, scan, or test the vulnerability of the service, its infrastructure, or its networks without explicit written permission.
- Attempt to breach, bypass, or circumvent multi-tenant workspace isolation boundaries, authentication mechanisms, or database access controls.
- Attempt to read, modify, delete, or infer the existence of another customer’s workspaces, runs, keys, files, or stored objects.
- Upload malware, viruses, trojans, or corrupted or hostile Git bundles and pack files designed to disrupt Git subprocesses, consume server memory, or execute unauthorized commands.
Abuse and resource exhaustion
Section titled “Abuse and resource exhaustion”- Circumvent or attempt to bypass plan quotas, rate limits, storage caps, body admission throttles, or concurrency limits.
- Automate mass account or workspace creation (for example, creating multiple accounts to evade the limit of one free workspace per user).
- Use disposable or fraudulent email addresses to register accounts.
- Launch denial-of-service (DoS) attacks, flood requests, or artificially consume shared infrastructure resources.
Unlawful and infringing content
Section titled “Unlawful and infringing content”- Store, transmit, or share material that infringes any copyright, patent, trademark, trade secret, or other intellectual property or proprietary rights.
- Transmit or store unlawful, fraudulent, defamatory, harassing, or harmful content.
- Publicly expose API keys (for example, by committing sync keys into public version control repositories or including keys in unprotected scripts).
3. Prohibited Data Classifications
Section titled “3. Prohibited Data Classifications”The service is a public preview and has no formal compliance certifications (such as SOC 2, ISO 27001, HIPAA, PCI-DSS, or FedRAMP). You must not upload or process:
- Protected Health Information (PHI): Data subject to HIPAA or similar health data regulations.
- Payment card and financial data: Credit card numbers, debit card details, bank account numbers, or credentials subject to PCI-DSS.
- Government identifiers: Social Security numbers, passport numbers, driver’s license numbers, or equivalent national identification numbers.
- Classified or export-controlled information: Material restricted under export control laws, defense trade regulations, or government security classifications.
4. Quotas and Free Tier Limits
Section titled “4. Quotas and Free Tier Limits”Users of the free preview plan must adhere to the configured service limits:
- Workspaces: Each account can own 1 free workspace at a time.
- Storage: 5 GB (decimal) of committed experiment run files and artifact objects per workspace.
- Daily uploads: 2 GB of uploaded data per UTC calendar day.
- Request rate: 600 requests per minute per workspace.
- People: 5 members and pending invitations per workspace.
When a workspace reaches its storage, upload, or member limits, requests return HTTP 403 Forbidden. When a workspace exceeds the request rate, requests return HTTP 429 Too Many Requests with a Retry-After header. Attempting to evade these limits violates this policy.
5. Enforcement and Suspension
Section titled “5. Enforcement and Suspension”We reserve the right to investigate potential violations of this policy. If we identify abusive activity, security threats, or terms violations, we may take immediate action, including:
- Suspending the workspace: The workspace is immediately disabled and answers HTTP
404 Not Foundon the dashboard, and sync clients are refused (HTTP403). - Revoking API keys and sessions: Blocking immediate access from automated sync clients or compromised credentials.
- Deleting the workspace: Permanently removing its data, and disabling the account’s access.
- Reporting to authorities: Reporting unlawful behavior to law enforcement when appropriate.
6. Reporting Violations
Section titled “6. Reporting Violations”If you discover a security vulnerability, evidence of abuse, or a violation of this policy, please report it immediately:
- Abuse and security reports: Email contact@agenticml.xyz.
- General technical issues: File an issue on GitHub Issues at agenticML/agenticml-lite/issues. For the hosted service, put “hosted” in the title.
7. Changes to This Policy
Section titled “7. Changes to This Policy”We may update this Acceptable Use Policy periodically. Continued use of the service following any updates constitutes acceptance of the modified policy.