Skip to content
KaamProof
How it worksProductPrivacyTermsContact
Open appJoin the pilot
How it worksProductPrivacyTermsContactOpen app

Legal

Privacy Policy

KaamProof helps agencies turn scattered project material into a reviewable, source-linked record. This policy explains what we collect, how we process your uploads in memory, and how we use AI only when you ask us to.

Effective: 10 Aug 2026 · Last updated: 10 Aug 2026 · Contact: hello@kaamproof.berlinxd.com

On this page1. Who we are2. What we collect3. How we process uploads4. AI processing (Meta AI)5. Pilot & contact data6. Storage & retention7. Sharing8. Security9. Your rights10. Cookies11. Children12. Changes13. Contact

1. Who we are

KaamProof is a project-record product operated on kaamproof.berlinxd.com. For privacy questions write to hello@kaamproof.berlinxd.com. We act as the data controller for the pilot form and as a processor for the project material you upload for analysis.

2. What we collect

Project material you provide

Files you upload (PDF, TXT, CSV, JSON, PNG, JPG, WEBP), text you paste, and the project name, client name and description you enter. We also receive technical metadata that browsers always send: IP address, user-agent and timestamps for rate-limiting and abuse prevention.

Pilot & contact submissions

Name, work email, agency name, team size and the free-text note about your workflow. This is only collected when you submit the pilot or contact form.

What we do not ask for

Do not upload passwords, payment card numbers, government IDs, or personal data about people who have not consented. The intake screen reminds you of this before you upload.

3. How we process uploads (in-memory, not stored)

This MVP processes uploaded files in memory only. Files are read during the /api/analyze request, sent to the AI provider if configured, and then discarded. We do not save your source files to disk or a database. The generated draft record lives only in your browser (in-memory) until you export it. If you close the tab, it is gone unless you downloaded it.

In short: Your source files are not stored on our server. Only the pilot and contact form entries are stored (see next section).

4. AI processing

When live analysis is enabled, project material is sent to Meta AI (muse-spark-1.2) (model set by META_MODEL, default muse-spark-1.2) via the Meta AI Responses API (api.meta.ai/v1) to extract approvals, scope changes, tasks, deliveries and warnings. We instruct the model to treat your material as untrusted data, not as instructions, and to return only factual, source-linked events.

If no MODEL_API_KEY (or META_API_KEY) is configured, live analysis is disabled and you can still use the sample project. Your material is never used to train models.

5. Pilot & contact data

Pilot submissions are stored as JSON lines in /opt/kaamproof/data/pilot.jsonl on the VPS and may be forwarded to the operator by email or internal tool. We use this only to contact you about the pilot, shape the product and respond to your message. You can ask us to delete it at any time.

6. Storage & retention

  • Uploaded project files: not stored — processed in memory per request, then dropped.
  • Generated draft record: stays in your browser; we never persist it server-side.
  • Pilot and contact entries: kept until you ask us to delete them or until the pilot concludes (max 12 months), then archived or deleted.
  • Server logs: standard nginx and Node access logs with IP and user-agent, rotated automatically.

7. Sharing

We do not sell your data. We share it only with (a) the AI provider (Meta AI) when you trigger analysis, (b) the hosting provider (Contabo VPS, EU), and (c) if legally required. The AI provider processes the request under Meta's API terms.

8. Security

Traffic is encrypted with TLS (Let's Encrypt). We set Content-Security-Policy, X-Frame-Options and Strict-Transport-Security, rate-limit /api/analyze (8 per 15 min) and pilot submissions (10 per 15 min), validate MIME types and size (max 10 MB per file, 5 files) and reject unsupported formats. No system is 100% secure — only upload material you are allowed to process.

9. Your rights

Depending on your location you may have the right to access, correct, delete or restrict processing of your data, and to object or ask for a copy. To exercise any right, email hello@kaamproof.berlinxd.com from the same email you used. We respond within 30 days. You also have the right to complain to your local supervisory authority.

10. Cookies & analytics

KaamProof does not set tracking cookies and does not load third-party analytics in this version. We use only essential browser storage (in-memory app state) to hold your draft record and a functional session if needed for the pilot flow. No advertising cookies.

11. Children

KaamProof is not directed at children and we do not knowingly collect data from them.

12. Changes

We may update this policy as the product evolves. We will change the "Last updated" date and, for material changes, highlight them on the site or by email if you are in the pilot.

13. Contact

Questions or requests? Email hello@kaamproof.berlinxd.com or use the contact page. For AI-related questions about how Meta AI processes your uploads, include the date and time of your analysis (from the app) so we can locate the server log entry (file names and sizes only — we do not retain file contents).

This policy is provided for transparency and is not legal advice.

KaamProof

Clear project history for agencies that work everywhere.

How it worksProductAppPrivacyTermsContact
© KaamProof. Built for clearer client work. hello@kaamproof.berlinxd.comPrivacy · Terms · Contact · AI suggests. Humans decide.