Terms of Service

These terms govern use of the SocialPipe API, operated by [to be confirmed: legal entity] ([to be confirmed: registered address]). By creating an account or using the API you agree to them.

1. The service

SocialPipe retrieves publicly available data from X (formerly Twitter) and returns it as JSON. X is the only platform we serve. We return the platform's own record without alteration — we do not select fields, rename them, or reshape the response. What a route returns is what the platform published about that post or account.

We do not host, store or redistribute video, audio or image files. Media addresses in a response point at the platform's own network; we return the address and never fetch the file.

2. Accounts and API keys

You are responsible for keeping your API keys secret and for all use made of them. A key is shown once at creation and cannot be recovered afterwards, because we store only a hash. Tell us immediately at support@socialpipe.dev if a key is exposed, and revoke it.

One account per organisation. Do not share keys with third parties or resell access to them.

3. Credits, billing and expiry

The API is priced in credits. Plans, credit allowances and per-call costs are on our pricing page.

• Cached responses are free. A repeat read of the same post inside its cache window returns cached: true, costs no credits, and leaves your balance unchanged
• Failed calls are not charged. Any response that is not a success leaves your balance untouched, and a call that fails after being charged is refunded
• Monthly credits expire at the end of the billing period they were granted for and do not roll over. One-time packs expire 12 months after purchase. Spending always draws from the soonest-expiring credits first
• Every plan is a hard stop, not overage. At a zero balance, requests are refused rather than billed. This holds on paid plans exactly as it does on the free tier: nothing bills you past the credits you bought, and there is no overage rate

Every credit movement is recorded in an append-only ledger. If our records and your expectations differ, that ledger is what we will reconcile against, and we will show you the entries.

[to be confirmed: legal entity] is the seller of record for your purchase. Card payments are processed by Stripe; we do not receive or store your card number. Prices are exclusive of VAT, GST and sales tax unless stated otherwise, and any such tax is added at checkout where we are required to charge it. [to be confirmed: tax registration status]

4. Cancellation and refunds

You may cancel at any time; access continues to the end of the paid period. We do not refund unused credits on cancellation. [to be confirmed: refund policy]

5. Acceptable use

You may not:

• Use the API to infringe copyright or other rights, or to circumvent a platform's technical protection measures
• Attempt to exceed rate limits by distributing traffic across accounts, or otherwise evade quotas
• Use the service unlawfully, or to harass, surveil or profile individuals
• Build or sell a product whose purpose is to profile, rank or score the individuals appearing in a response, or to combine that data with other sources for that purpose
• Use twitter/followers, twitter/following, twitter/retweeters, twitter/comments or twitter/comments-latest to assemble a social graph about identified people — to map who a named person associates with, to infer things about them from who follows them, or to build a list of individuals selected by who they follow or what they replied to. Retrieving a page to count it, rank the posts in it, or measure an audience in aggregate is fine. Retaining the individuals as a list of people is what this prohibits
• Derive or record a special category of data about an individual from a response — racial or ethnic origin, political opinions, religious or philosophical beliefs, trade union membership, health, sex life, sexual orientation — or their immigration status. A follower graph is a common way to infer several of these, which is why it is written down here rather than assumed

On the bulk routes, and why this clause is the real control. Five of our twelve routes return pages of people who are not the subject of the request. We cannot inspect what you build, and we do not want the ability to; the honest position is that the terms are the control and enforcement is complaint-driven. That is weaker than a technical restriction and we would rather say so than imply otherwise. If you need bulk graph data for something the list above prohibits, this is not the right supplier.

On reselling. An earlier version of these terms prohibited reselling raw API responses as a competing extraction service. That clause is withdrawn: SocialPipe is itself a reseller of an upstream data provider, and a prohibition we do not hold ourselves to is not one we intend to enforce. Build what you like on top of these responses, including something that competes with us. What the clause above still prohibits is the use we will not support at any layer — profiling the people in the data.

The prohibition on re-identifying hashed data is also withdrawn, because nothing is hashed any more. See §7 for what that means for your obligations.

We may suspend an account for a breach of this section. Where the breach is a copyright matter, the process in our DMCA policy applies, including termination after three actioned notices in a rolling twelve months.

6. Availability

We aim for high availability but do not promise uninterrupted service on any plan unless a separate written service level agreement says otherwise. [to be confirmed: sla terms]

Upstream platforms change without notice and may block or degrade extraction. Where we detect that a platform is returning unusable results we return an explicit error rather than charging for empty data. Interruption caused by an upstream platform is not a failure of these terms.

7. Third-party platforms, and the people in the data

SocialPipe is not affiliated with, endorsed by, or sponsored by any platform whose public data it retrieves. Your use of that data is your responsibility, including compliance with the source platform's terms and with copyright law in your jurisdiction.

A response contains personal data about people who are not our customers and not yours. Because we return the platform's record unchanged, that record typically includes handles, display names, biographies, profile images and engagement counts — for the account you asked about, and on a timeline for anyone they quoted or mentioned. Nothing in a response is pseudonymised, hashed or redacted.

Five routes return people in bulk, and they are the ones to think about before you call them. twitter/followers, twitter/following and twitter/retweeters return pages of accounts — dozens to hundreds per call — none of whom is the account you named. twitter/comments and twitter/comments-latest do the same for the authors of replies. One call to any of them is a social graph or a conversation, not a lookup, and paging through one assembles a dataset about identified people that neither they nor the subject of your query consented to. Section 5 limits what you may do with it; this section is the notice that you are receiving it.

Your own call history. GET /v1/usage returns the calls you made — route, credits, whether it was cached, and when. Where a deployment has enabled it, that history can also include the parameters you sent and the response you received. It is scoped to your account and no other customer can read it. Two things follow: a response you fetched once may still be readable through us after X has deleted the original, and the third-party personal data in it is subject to section 5 in that history exactly as it was on the day you fetched it. The Privacy Policy says whether that storage is switched on.

If you store it, you are the controller of it. Retrieving public data through our API does not make us responsible for what you do with it afterwards. You need your own lawful basis for retention, profiling or enrichment, and your own answer for a data subject who asks you what you hold. That applies with most force to the five routes above: a stored follower graph is a controller-side database of people, and a data subject's rights against it are yours to answer, not ours. We hold a short-lived cache and nothing else; see the Privacy Policy.

8. Data protection

Our handling of personal data is described in the Privacy Policy and the subprocessor list. A data processing agreement is available on request at privacy@socialpipe.dev.

9. Intellectual property

We own the API, documentation and software. You own your account data and the use you make of the output, subject to the rights of whoever owns the source material. We grant no rights in third-party content, and none are implied.

10. Warranties and liability

The service is provided "as is" to the fullest extent permitted by law. We do not warrant that extracted data is complete, accurate or fit for a particular purpose.

[to be confirmed: liability cap]

Nothing in these terms limits liability that cannot lawfully be limited, including for death or personal injury caused by negligence, or for fraud.

11. Changes

We may change these terms. Material changes are published here with a new version and effective date and notified to account holders at least [to be confirmed: change notice period] before taking effect. Continuing to use the API after that is acceptance.

12. Governing law

These terms are governed by the laws of [to be confirmed: governing law], and the courts of [to be confirmed: jurisdiction] have exclusive jurisdiction.

13. Contact

support@socialpipe.dev · [to be confirmed: legal entity], [to be confirmed: registered address]