Effective date: 14 August 2026
By using Opafex you agree not to build, deploy, or use an agent to do any of the following. This protects you, the people your agent interacts with, and our standing with our AI and infrastructure providers.
1. Prohibited high-risk and unlawful uses
You must not use the Service for, or to build agents that perform, automated decision-making in these areas without appropriate human oversight, legal basis, and (where applicable) EU AI Act high-risk compliance — and in several cases not at all:
- Prohibited outright: social scoring; manipulative or deceptive techniques that distort behaviour; exploiting vulnerabilities of specific groups; real-time remote biometric identification; biometric categorisation by sensitive traits; untargeted scraping of facial images.
- High-risk (not permitted on the standard Service): decisions about creditworthiness, insurance, employment/recruitment, worker management, education or exam scoring, access to essential public or private services or benefits, law enforcement, migration/border control, or administration of justice.
- Unlawful generally: anything illegal in the applicable jurisdiction, infringing IP, defamatory, or that facilitates fraud, harassment, or violence.
2. Data and privacy obligations
- Have a lawful basis and provide required notices for any personal data your agent processes.
- Do not upload special-category data (health, biometric, etc.) unless you have a valid basis and have told us.
- Do not use the Service to process children's data in violation of law.
- Honour the rights of individuals your agent contacts (e.g. unsubscribe, do-not-contact).
3. Communications and outreach
- Comply with anti-spam and marketing law (UK PECR, EU ePrivacy, US CAN-SPAM/TCPA) for any email, SMS, or voice your agent sends. Obtain consent where required, identify yourself, and include opt-out.
- Do not send unsolicited bulk messaging, spoofed sender identities, or robocalls in breach of law.
4. Security and integrity
- No attempts to bypass authentication, access other tenants' data, probe for vulnerabilities without authorization, exfiltrate secrets, or overload the Service.
- No using the Service to develop malware, conduct DoS, or compromise third-party systems.
- Web/browser automation must respect target sites' terms and robots directives; only public content.
5. AI-specific
- Do not present AI-generated output as human where the law requires disclosure.
- Do not use outputs to train a competing model in breach of our or our providers' terms.
- You are responsible for reviewing agent outputs before relying on them for consequential actions; sensitive actions are gated behind human approval for this reason.
6. Enforcement
We may suspend or terminate access for violations, and must report certain illegal content/conduct as required by law. We will use proportionate measures and, where possible, give notice and a chance to cure.
Report abuse: abuse@opafex.com.