dataloup NosyPup
NosyPup dog mark
Privacy policy

NosyPup

NosyPup page·Support·[email protected]

Effective: 21 October 2026 · Operator: dataloup, Abu Dhabi, UAE · Contact: [email protected]

Nosy Pup is a social-listening tool. It finds public posts that match keywords a developer chooses, and shows them to that developer so they can decide whether to reply by hand. This policy explains what it collects, why, and what happens to it. It is written to be read, not skimmed.

1. Who this applies to

Two groups of people:

  • Users — developers who run Nosy Pup for their apps. NosyPup subscriptions are available to customers.
  • Authors — people whose public posts on Threads, Bluesky, Reddit or the App Store match a user’s keywords. Authors have no account with us and never interact with Nosy Pup directly.

2. What we collect

From authors (public content only)

When a public post matches a keyword, we store:

  • the post text and title, as published
  • the author’s public handle or display name
  • the post’s link, timestamp and platform
  • public engagement counts (likes, replies, reposts) where the platform provides them
  • for App Store reviews of a user’s own app: the review text, star rating, reviewer nickname and territory

We collect only content that is already public. We do not access private accounts, direct messages, follower lists, email addresses, locations or any data behind a login other than the user’s own.

From users

  • Account identifiers needed to run the service: shared sign-in account and workspace details, subscription and entitlement references, optional platform credentials stored encrypted, a Telegram user ID for linked alerts, and the keywords and app descriptions the user enters. Payment card details are handled by the payment provider, not stored in NosyPup.
  • Records of what the user did with each post: claimed, skipped, replied, and the text of any reply the user drafted.
  • Click counts on the user’s own campaign links (country and browser type only — no cookies, no cross-site tracking).

What we do not collect

We use authentication cookies to keep users signed in. We do not use advertising cookies or advertising identifiers. No tracking pixels. No profiles of individual authors. No sale or licensing of any data to anyone.

3. How we use it

  • Triage. Relevant public post content and app context may be sent to Anthropic’s Claude API to assess relevance and draft a suggested reply. This processing is subject to Anthropic’s API terms.
  • Alerts. Users can review discoveries in the browser and receive alerts through Telegram when linked.
  • Human control. Users can copy a draft and publish it themselves. Optional connected posting on supported platforms requires the user to connect an account and approve the reply. Saving a draft or follow-up does not publish it or send an automatic message.
  • Attribution. Campaign-link clicks and user-recorded outcomes help users review activity. They do not prove that a reply caused an install.

4. Platform data

  • Threads (Meta). Public post data is accessed through the Threads API under Meta’s Platform Terms and Developer Policies. Available access depends on the connected account and granted permissions. We do not store Threads data beyond the retention period below, and we honour deletion requests within 10 days.
  • Bluesky. Public posts via the AT Protocol, authenticated as the user’s own account.
  • Reddit. Public conversation links can arrive through supported monitoring alerts. Users review the original conversation and publish any reply under Reddit’s rules. Dataloup handles monitoring setup.
  • Apple App Store. A user’s own app reviews via the App Store Connect API, using the user’s own credentials.

5. Where it lives

Data is hosted using Cloudflare Workers, D1 and KV. This policy does not promise exclusive EU storage. Sensitive integration credentials are encrypted; account owners should not include credentials in support screenshots or messages.

6. How long we keep it

  • Stored discoveries, replies, drafts and activity records support the workspace history while the account is active.
  • The scheduled cleanup removes triaged or skipped discoveries older than 730 days only when they have no linked posting attempt, evaluation label or click. Other records may remain longer; there is no blanket 90-day automatic deletion policy.
  • Users and authors can request deletion as described in section 8. Do not rely on a shorter automatic deletion period for saved content or click records.

7. Who we share with

  • Anthropic (triage, as above), Cloudflare (hosting), Telegram (alert delivery), Stripe (subscription payments through the shared billing service), and Apple, Meta and Bluesky as the platforms the data comes from — each only to the extent needed to provide the service.
  • No advertisers, no data brokers, no resale, ever.
  • We will disclose data if legally required and will tell the affected user unless prohibited.

8. Your rights

Authors: if a public post of yours is stored by Nosy Pup and you would like it removed, email [email protected] with the post link. We delete it within 10 days, no questions.

Users: you can export or delete everything we hold about you by emailing the same address. Deletion closes the account and removes all stored posts, replies and tokens within 10 days.

If you are in the EU/UK you have the rights set out in the GDPR, including access, correction, erasure, restriction and portability, and the right to complain to your data-protection authority. We act as the data controller for the data described here.

9. Children

Nosy Pup is a professional tool for developers and is not directed at anyone under 18.

10. Changes

We will post changes here with a new effective date. Material changes will be announced to users in their alert channel before they take effect.

11. Contact

[email protected] · dataloup, Abu Dhabi, United Arab Emirates