Trust center
How we protect the platform and your data, who processes it, how we keep scraping responsible and how to report a security issue.
Security overview
The controls below are in place today. We do not hold SOC 2 or ISO 27001 certification. If you need answers for a security review, write to [email protected].
Encryption in transit
Every public endpoint is HTTPS only, with TLS 1.2 or newer and HSTS. Our servers have no open inbound ports: traffic reaches them through an encrypted tunnel from our edge network.
Encryption at rest and backups
Backups, stored results and log archives live in encrypted object storage in the EU. The database is archived continuously, analytics and the secrets vault daily. Restoring all three is rehearsed and timed.
Signed images only
Production runs only container images built by our CI, pinned by digest and signed with our key. An admission controller rejects everything else.
Least-privilege networking
Each service runs in its own namespace with network policies that allow only the connections it needs. Scraping workers cannot reach our databases or the internet directly, only the proxy layer.
Secrets in a vault
Credentials are kept in a self-hosted secrets vault and synced only to the service that uses them, never into source code. Administrative access to the vault needs a password and a one-time code.
Two-factor authentication for admins
Accounts with admin rights must use two-factor authentication. Internal tools sit behind an identity-aware proxy and are not reachable from the open internet.
API keys stored as HMAC hashes
We store only an HMAC-SHA256 of each API key, keyed with a secret held in the vault rather than the database. We cannot read your key back, and a revoked key stops working within a minute. Passwords are stored hashed.
EU company, EU hosting
Siftzo is an EU company. The platform, its databases, backups and logs are hosted in the EU.
Data handling
We collect what we need to run and bill the service, keep it in the EU and delete it on a schedule. The privacy policy is the binding version; this is the short form.
What we record about API requests
For each request we record the time, API key ID, target domain, status, tier, credits, country and duration, plus the shape of the path with numbers, IDs and the query removed (so we can see which kind of page fails). That is what billing, your usage dashboard and abuse handling need. These records hold the domain, not the full URL, unless your organization turns on full URLs for its request log; those are kept for 48 hours. They never hold the page content.
How long we keep it
- Results of sync requests are passed straight to you and not stored.
- Results of async requests are deleted after 24 hours, crawl results after 7 days.
- Request records are kept for up to 400 days, full URLs (where turned on) for 48 hours, application logs for 30 days.
- Account data is kept while the account exists. After deletion it remains in encrypted backups for up to 30 days.
- Billing records are kept as long as bookkeeping law requires.
Content you retrieve
Pages you fetch may contain personal data. We process that content only to deliver it to you and never use it for our own purposes. A data processing agreement is available on request at [email protected].
No tracking
The public website sets no cookies and runs no analytics. The dashboard uses only the cookies it needs to keep you signed in. Our metrics, logs and traces stay on our own systems in the EU.
Sub-processors
These companies process data on our behalf under data processing agreements. The list is the same as in the privacy policy. We tell customers before adding or replacing one.
| Provider | Purpose | Location |
|---|---|---|
| Akamai Technologies (Linode) | Hosting and infrastructure | EU |
| Cloudflare, Inc. | Network protection, DNS, bot checks and encrypted backups | Global network; stored data in the EU |
| Stripe Payments Europe, Ltd. (Ireland) | Payments and invoicing (paid plans only) | EU entity |
| Mailgun Technologies, Inc. | Account and service emails | Sent from its EU region |
| Decodo (Lithuania) | Residential proxy provider that carries requests to target websites. It sees the target URL but not your identity or account details | EU entity |
| Anthropic, PBC | AI extraction only: the page content and your schema or prompt for /v1/extract and the Data API's AI fallback. Not used to train models | United States (standard contractual clauses) |
Responsible scraping
- robots.txt in crawls. The Crawl API always follows robots.txt and its
Crawl-delay, and this cannot be turned off. Pages that robots.txt disallows are skipped and not billed. - A domain blocklist. Government and banking sites, and sites whose owners have asked to be excluded, are refused before any traffic is sent.
- Public targets only. Private networks, internal hostnames and non-public IP addresses are refused.
- Clear rules. The acceptable use policy forbids credential stuffing, bulk account creation, overloading sites and collecting personal data without a legal basis. Accounts that break it are suspended.
Site owners can report abuse or ask to be excluded at [email protected]. Please include the domain, the time range and example requests if you have them. More in Responsible use.
Vulnerability disclosure
If you find a security issue in Siftzo, please tell us at [email protected]. Our contact details are also published in security.txt.
- Describe the issue, the affected URL or endpoint and how to reproduce it.
- Test only against your own account and data. Do not access, change or delete other customers' data, and stop and tell us if you come across any.
- Do not run denial-of-service tests, spam or social engineering, and do not use the API to attack third-party sites.
- Give us reasonable time to fix the issue before you disclose it publicly.
We will confirm that we received your report, keep you updated while we work on it and credit you if you want. We do not run a paid bug bounty.
Status
Live status and 90-day uptime for the website, dashboard and every API, checked every minute from outside our own infrastructure.
View status