Acceptable Use Policy

1. Scope

Web Answerly is operated from Victoria, Australia. The operating company is being registered. Its legal name, ABN and ACN will appear here, and it will be the party to this document, before the document takes effect. In this document "Web Answerly", "we", "us" and "our" mean that company.

This policy applies to every customer, team member and API user, and to every agent a customer deploys. You are responsible for how your agents are configured and used, including by your End Users, as far as you can reasonably control it. Terms used here have the meanings given in our Terms of Service.

2. Prohibited outright

You must not use Web Answerly, or deploy an agent, to:

• make significant decisions about people without human review: using answers, actions or tools to decide, by themselves, a person's access to employment, credit or lending, housing, insurance, healthcare, education or government benefits, or any other decision with legal or similarly significant effects;
• provide clinical or emergency services: diagnosis, treatment or triage, or acting as a primary source of help in an emergency or crisis. An agent may give general information, and should direct people to appropriate services;
• target children: deploy an agent directed primarily at people under 18;
• break the law, or help anyone else do so;
• defraud, deceive or manipulate people, including by impersonating a person or organisation, presenting an agent as a human, or hiding or contradicting the widget's notice that answers are generated by AI;
• harass, threaten, abuse or discriminate against anyone, or generate or distribute content that is unlawful, defamatory, sexually exploitative or that incites violence or hatred;
• infringe intellectual property, privacy, confidentiality or other rights, including by indexing content you have no right to use;
• send spam, run phishing, or distribute malware;
• monitor, profile or track individuals covertly, or build surveillance tools.

3. Crawling and content

• Submit only websites you own or are authorised to have crawled, and connect only documents and accounts you are authorised to share with us.
• Do not use the crawler to get around logins, paywalls, access controls or robots.txt rules, to collect content for purposes unrelated to your agent, or to load a site heavily.
• Do not index or upload sensitive information an agent should not disclose, such as health records, payment card data, government identifiers, passwords or other secrets. Assume anything you index can be repeated in an answer.

4. Attacking or abusing the Service

You must not:

• try to break workspace isolation, gain access you were not given, or test our security without our written permission (we welcome responsible reports — see section 8);
• use prompt injection or manipulated content to make an agent, ours or another customer's, reveal its instructions or data, ignore its rules, or attack third parties;
• get around rate limits, plan limits, usage allowances, verification checks or abuse protection, including by creating accounts or workspaces to obtain extra free usage;
• overload the Service, or flood chat, lead, booking or demo forms, APIs or the crawler with automated traffic;
• offer the Service's AI models to others as a general-purpose model API, or copy the Service to build a competing product, without our written agreement (running agents for your own clients, as an agency does, is permitted);
• use the Service where sanctions or export laws prohibit it.

5. Actions, API actions and MCP tools

• Point actions only at systems you operate or are authorised to use, and give them only the permissions the agent needs.
• Require confirmation for actions that change something important, such as payments, refunds, account changes or deletions, and do not configure an action to cause an effect the End User has not asked for.
• Do not use actions to reach internal networks, cloud metadata services or other systems you are not authorised to access. We block these addresses; trying to get around the block is a breach of this policy.
• Approve an MCP tool only after you have reviewed what it does, and review it again when it changes.

6. Regulated and high-stakes topics: human oversight required

An agent may give general information on medical, legal, financial, tax, insurance, immigration, employment and other regulated topics only if:

• it is not presented as personalised professional advice, and End Users are directed to a qualified person for advice about their own situation;
• a qualified person in your organisation has reviewed the content the agent draws on, and you monitor its answers;
• you use the controls available — trust rules, verified answers, fallback messages and human handoff — to restrict what it says on those topics; and
• you meet any licensing, disclosure or record-keeping rules that apply to you.

7. Information End Users should not share

The widget asks visitors not to share passwords, card details or other secrets. Do not design an agent, form or action that asks End Users for payment card numbers, bank or account passwords, government identification numbers, or health information, unless we have agreed to that in writing.

8. Reporting and enforcement

• Report abuse, or an agent you believe breaches this policy, to support@webanswerly.com. Report security vulnerabilities to support@webanswerly.com.
• If you breach this policy we may remove content, disable an action, key, agent or website, or suspend or close the workspace, as set out in the Terms. Where it is lawful and practical, we will tell you first and give you a chance to fix the problem. We may act immediately if there is a risk of serious harm.
• We may report unlawful activity to the relevant authorities.