Privacy Policy
SciCoagent ("we", "us") builds research tools for scientists. This policy explains what information we collect, how we use it, who we share it with and the choices you have. Sections 1–10 apply to every SciCoagent product. Section 11 adds the details specific to each product. Currently those products are:
- Benjamin (benjamin.scicoagent.com): AI proofreading for scientific manuscripts.
- Eratosthenes / SciCo-Search (search.scicoagent.com): AI-assisted scientific literature search.
When we launch a new product, we will add it to Section 11 before it collects any personal information.
1. One account across products
All SciCoagent products use one sign-in system. When you sign in with Google on any of our products, the same account is used across them. Subscriptions, plans and usage limits are kept separately for each product. A subscription to one product does not give you, or charge you for, another.
2. Information we collect
- Account information. When you sign in with Google, we receive your email address, display name and profile picture URL. We use them to sign you in, identify your account and show who is signed in.
- Content you submit. This means what you give a product to work on, such as documents for proofreading or search queries. Section 11 describes what each product does with it.
- Plan, usage and billing records. For each product, these are your plan, your usage against that plan's limits (for example words proofread or searches run), and your subscription status, period and renewal date. Payments are handled by our payment partners. We never see or store your full card details.
- Technical and usage data. These are basic server logs (such as IP address, browser type, time of request and errors) and, where enabled, product analytics (see Section 6).
- Messages you send us. If you email us, we keep the correspondence so that we can help you.
3. How we use information
- to provide each product, apply its plan limits and process its subscriptions;
- to contact you about your account, billing, security or important changes;
- to keep the products secure, prevent abuse and fix problems;
- to understand, in aggregate, how the products are used so that we can improve them.
We do not sell personal information, and we do not use it for third-party advertising.
4. AI processing
Our products use AI models to do their work. Where a product sends your content to an AI provider, Section 11 says so. Today that provider is Google, through the Gemini API. Google processes that content under the Gemini API Additional Terms of Service. Under those terms, the handling of submitted content depends on the service tier used. For unpaid tiers, Google may use content to improve its products, and human reviewers may read it. For paid tiers, Google does not. We do not use your content to train any model of our own. Please do not submit material you are not permitted to share with a cloud AI provider, for example embargoed, classified or regulated personal data.
5. Service providers and sharing
We share information only with the providers we need to run our products, and only for that purpose:
- Google: Google sign-in and the Gemini API (AI processing).
- Supabase: authentication, account records, plans, usage counters and saved data.
- Hugging Face: hosting of our application servers.
- GitHub Pages: hosting of our websites.
- Microsoft Clarity: website analytics, where enabled (see Section 6).
- Dodo Payments (international) and Razorpay (India): payment processing for paid plans.
We may also disclose information if the law requires it, to protect our rights or users' safety, or as part of a merger or acquisition. In that case, this policy continues to apply to the information transferred.
6. Cookies, local storage and analytics
- Sign-in. We store your sign-in session in your browser's local storage so that you stay signed in.
- Preferences. Products may store small conveniences in your browser, such as display settings.
- Analytics. Some of our pages use Microsoft Clarity to understand how visitors use them. Clarity uses cookies and records interactions such as clicks, scrolling and page views. It also produces session replays, in which text you type into forms is masked. Microsoft processes this data under the Microsoft Privacy Statement. You can block these cookies in your browser settings without affecting how the products work.
We do not use advertising cookies.
7. Retention
- Account and plan records are kept while your account exists.
- Content you submit is kept as described for each product in Section 11.
- Billing records are kept for as long as tax and accounting law requires, even after you delete your account.
- Server logs are kept for a limited period for security and debugging, then deleted.
8. Your choices and rights
You can ask us to access, correct, export or delete your personal information, or object to how we use it, by writing to contact@scicoagent.com from your account email. Deleting your account removes your account, plan and usage records and your saved content in every SciCoagent product. Billing records that we must keep by law are the exception. You can also revoke SciCoagent's access to your Google account at any time at myaccount.google.com/permissions. Depending on where you live, you may also have the right to complain to your local data-protection authority.
9. Google user data
When you sign in with Google, we request only basic profile scopes: your email address, name and profile picture. We use this data only to sign you in and operate your account, as described in this policy. We do not use it for advertising, sell it, or let humans read it except for security, legal compliance or support you ask for.
SciCoagent's use and transfer of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements.
10. Security, international transfers, children and changes
- Security. We use encrypted connections (HTTPS) and access controls, and we limit who can access data. No system is perfectly secure. If a breach affects you, we will tell you as the law requires.
- International transfers. Our providers may process data in countries other than yours, including the United States, the European Union and India. Where required, those transfers rely on the providers' standard safeguards.
- Children. Our products are for researchers and are not directed at children under 16. We do not knowingly collect information from them.
- Changes. We may update this policy. The effective date at the top shows the current version. For material changes, we will notify signed-in users by email or in the product.
11. Product-specific details
Benjamin: benjamin.scicoagent.com
Documents you upload. These are the manuscripts you submit (LaTeX projects, Word, plain text, Markdown or RTF) and the corrected output. To proofread a document, its text is sent to the Gemini API (see Section 4).
Retention of documents. Documents are processed transiently. They are held only in temporary server storage while your run and download are available. They are not kept in a database, and they are removed automatically, at the latest when the server restarts. They are not shared with other users.
Usage. We record the words proofread each month to apply your plan's allowance. We do not record the text itself.
Authors of public arXiv papers. To demonstrate the service, we may proofread papers that are publicly available on arXiv. We then send one email to the author addresses published in the paper, with the suggested corrections attached. We store only the address and the date it was contacted, so that nobody is contacted twice. Reply "unsubscribe" or write to us, and we will never contact that address again.
Gmail. Benjamin's outreach tool uses the Gmail API scope gmail.send only on the Benjamin team's own
Gmail account. It uses that scope to send the one-time author emails described above, after a team member approves each one. It can
only send messages. It does not read, list, modify or delete any email, it never accesses any user's or recipient's Gmail
account, and no Gmail data is stored or shared.
Eratosthenes / SciCo-Search: search.scicoagent.com
Search queries. These are the queries you enter and the options you choose. To find and rank papers, your queries are sent to OpenAlex, an open scholarly index. Queries and the titles and abstracts of retrieved papers may also be sent to the Gemini API (see Section 4) to expand queries, score relevance and power analysis features.
Saved sessions. Your search sessions (the query, the results and the analysis) are saved to your account so that you can restore them later. They are kept until you delete them or your account.
Usage. We count feature use per month (for example deep searches) to apply your plan's limits.
Teams. If you join a team plan, the team owner can see that you are a member and your account email. Your saved sessions remain your own unless the product clearly offers, and you choose, to share them.
Paper data. Bibliographic information about published papers and their authors comes from public scholarly sources and is shown as part of search results.
12. Contact
Questions or requests about privacy: contact@scicoagent.com.