Apify Confirms Data Breach After Credential Theft

A compromised credential let an attacker into Apify's internal systems on Nov 9-10, exposing some customers' data before the company locked it down.

Apify, the web-scraping and automation cloud platform used by an estimated 91,000 customers, has confirmed that an unauthorized third party used a compromised credential to access parts of its internal systems between November 9 and 10, 2026, and that some customers' personal information and stored data were accessed as a result. The company detected the intrusion on November 11, cut off the attacker's access, revoked a batch of OAuth connections including GitHub, and began emailing affected customers individually. By November 13, Apify said it had moved to storing API tokens only as hashes and was adding new network and authentication controls, with a full post-mortem promised within a week.

Here is what is confirmed so far, what Apify says it is doing about it, and what you should do right now if you run Actors, store data, or connect integrations on Apify.

Key facts
  • What happened: An attacker used one compromised credential to reach parts of Apify's internal systems and pivot into related accounts.
  • When: Intrusion window Nov 9-10, 2026; detected and contained Nov 11, 2026.
  • Who is affected: A subset of Apify's roughly 91,000 customers; Apify is notifying affected users directly by email.
  • What Apify did: Closed the access path, revoked some OAuth connections (including GitHub), rotated internal secrets, hashed all stored API tokens, added network restrictions, and notified authorities.
  • What's not confirmed by Apify itself: The detailed list of exposed data categories (passwords hashes, 2FA recovery codes, phone numbers, partial card data) comes from a customer-shared breach notification email reported by SecurityWeek, not from Apify's public status page text.

Who is Apify, and why this incident is a big deal

Apify is a Prague-founded cloud platform for web scraping and browser automation. Developers and businesses use it to build and run "Actors" — containerized scraping and automation programs — on Apify's infrastructure, store the scraped output in Apify's managed datasets and key-value stores, and schedule the whole pipeline to run on autopilot. The company says it serves roughly 91,000 customers, ranging from individual developers automating a side project to companies that depend on Apify-run jobs for price monitoring, lead generation, AI training-data collection, and other data-pipeline work. For background on the underlying practice, web scraping itself is a widely used and largely legal technique, though it increasingly intersects with privacy and platform-terms questions — which is part of why a platform built around it, holding other people's API keys and scraped data at scale, is a meaningful target.

That scale and position in the pipeline is exactly what makes this incident more consequential than a typical SaaS outage. An Apify account is rarely self-contained: it commonly holds a GitHub connection (to pull Actor source from private repos), a Google or GitHub OAuth login, API tokens used by other internal tools to trigger Apify runs, and datasets that may contain scraped personal information belonging to third parties who never signed up for Apify at all. A compromise of Apify's internal systems therefore has the potential to touch a wider blast radius than the credentials of Apify's own 91,000 customers.

What happened in the Apify security incident

According to Apify's own status page, an unauthorized third party gained access to parts of the company's internal systems on November 9 and 10, 2026. Apify's security team detected the unusual activity on November 11 at 12:13 UTC, cut off the access path, and opened an investigation. The company says a single compromised credential allowed the attacker to obtain additional account credentials and reach related accounts — a classic credential-pivoting intrusion rather than a software exploit tied to any specific CVE.

As an immediate containment step, Apify revoked some OAuth connections to external integrations, including GitHub, and said it would prompt users of other affected integrations to reconnect by email. The company also added monitoring and began rotating internal secrets. Apify states that as of its November 12 update, it had found no further attacker activity after November 11 and that the access path responsible for the intrusion was closed.

Apify breach timeline

DateDevelopment
Nov 9-10, 2026Attacker uses a compromised credential to access parts of Apify's internal systems.
Nov 11, 2026Apify detects the activity (12:13 UTC), blocks further access, revokes some OAuth connections including GitHub, and opens an investigation.
Nov 12, 2026 (AM)Apify confirms the compromised credential led to further account access; rotates critical internal secrets; notifies relevant authorities; tells users to reconnect GitHub/other integrations if prompted.
Nov 12, 2026 (PM)Apify confirms some customers' personal information or stored data was accessed and begins emailing affected customers individually.
Nov 13, 2026Apify says API tokens are now stored only as hashes, adds network restrictions and stronger internal authentication, advises users to set storage access to "Restricted," and promises a post-mortem within a week.

What data was reportedly exposed

Apify's own public status-page updates confirm, in general terms, that "some customers had their personal information or stored data accessed during the attack," without publishing an itemized list of data categories. A more detailed breakdown has circulated from a different source: SecurityWeek reported on the contents of a breach-notification email that an affected customer shared, which listed:

  • Names, email addresses, billing details, account activity logs, and IP addresses
  • Salted password hashes, two-factor authentication recovery codes, and phone numbers
  • Authentication tokens tied to Google and GitHub sign-ins
  • Limited payment card data — cardholder name, expiration date, and the last four digits only

It's important to be precise about sourcing here: that itemized list comes from a customer-shared notification email as relayed by SecurityWeek, not from language Apify has published on its own status page. Treat it as a credible but secondary account of what an individual notification said, rather than as Apify's own public confirmation of every category for every affected account.

Reportedly exposed (per customer notification cited by SecurityWeek)Explicitly NOT exposed (per the same report)
Name, email, billing info, IP addresses, account activityFull payment card numbers
Salted password hashesCard security codes (CVV/CVC)
2FA recovery codes, phone numbers 
OAuth tokens linked to Google/GitHub sign-in 
Cardholder name, card expiration date, last 4 digits 

SecurityWeek also noted that Apify's investigation found no evidence that the revoked OAuth connections themselves were an attack vector into customer systems — the connections were cut as a precaution, not because they were confirmed to be how the attacker moved.

How Apify responded

Beyond the immediate containment on November 11, Apify's November 13 update describes several concrete hardening steps:

  • Token storage changed: API tokens are now stored only as salted/hashed values rather than in a retrievable form, which Apify says limits the blast radius of a similar future incident.
  • Network and auth hardening: New network restrictions between internal services and stronger authentication requirements on internal systems, intended to close the specific paths the attacker used.
  • Ongoing review: A security review of the full stack, continued log analysis, and monitoring for any further suspicious activity.
  • Customer guidance: Apify is telling users to set storage access to "Restricted" in the Apify Console, because anyone who knows a storage's ID can read an unrestricted storage by default.
  • Authorities notified: Apify says it has notified relevant authorities, consistent with breach-notification obligations in jurisdictions where affected customers are based.

Apify has committed to publishing a full post-mortem, including root cause and remediation detail, within a week of the November 13 update. As of this writing, the incident remains open on Apify's status page, with Actors and the Apify API still flagged for partial impact while the review continues.

Why this matters for Apify users

Apify sits in an unusually sensitive spot in a lot of customers' stacks: it runs scheduled Actors (scraping and automation jobs) that often hold API keys, proxy credentials, and OAuth grants to other services, and it stores the output of those jobs — sometimes including scraped personal data, internal documents, or credentials incidentally captured during a run. A credential-based intrusion into Apify's internal systems is therefore not just "Apify's problem" — it can cascade into every service a customer connected to their Apify account, from GitHub repos holding Actor source code to Google accounts used for sign-in.

That's also why this incident reads differently than a typical unpatched-CVE story: there's no patch to apply, because the entry point was a stolen credential rather than a software flaw. The defenses that matter most here are the account-hygiene basics — unique credentials, rotated secrets, reviewed OAuth grants — rather than a version bump.

What Apify users should do now

Whether or not you've received a direct notification from Apify, security teams generally recommend treating an incident like this as a prompt to review exposure rather than waiting:

  1. Read any email from Apify carefully and follow its specific instructions — Apify says it is contacting affected customers individually, and the email may reference account-specific detail.
  2. Reconnect GitHub (and other) integrations if you build Actors from private repositories. Apify revoked some OAuth connections as a precaution; go to your integration settings, use "Change provider," and re-authorize through GitHub (or the relevant provider) rather than assuming the old connection still works silently.
  3. Rotate your Apify API token(s) and account password. Even though Apify says tokens are now stored only as hashes going forward, rotating any token that existed before the hashing change removes any residual risk from the plaintext/retrievable period.
  4. Review OAuth grants on connected accounts (Google, GitHub, and any other service you've linked to Apify) and revoke anything you don't recognize or no longer use.
  5. Set storage access to "Restricted" in Apify Console under Settings → General resource access. Apify has flagged that an unrestricted storage can be read by anyone who knows its storage ID.
  6. Enable or re-check two-factor authentication on your Apify account, and treat any previously issued 2FA recovery codes as potentially compromised if you were notified as an affected customer.
  7. Watch for phishing. Apify has explicitly said it will never ask for your password, 2FA codes, or payment details — treat any message that does as a scam, even if it references real incident details.
  8. Check billing statements for unrecognized charges if you were told payment-related fields (cardholder name, expiration, last four digits) may have been exposed, and ask your card issuer about reissuing the card if you have any doubt.

Why a "compromised credential" incident is different from a CVE

Much of the security news cycle in any given month is dominated by a specific flaw — a CVE added to a vulnerability catalog, a vendor shipping a patch, defenders racing to apply it before attackers scan for unpatched systems. The Apify incident is a different animal, and it's worth being explicit about why. Apify has not disclosed a software vulnerability, and there is no CVE identifier attached to this event. Instead, the company's own account describes an attacker who obtained one valid credential — Apify has not said whether that credential belonged to an employee, a contractor, or a third-party service account — and used it to reach internal systems, then pivoted from there into other credentials and accounts.

That distinction matters for how defenders should respond. A CVE-driven incident is usually resolved by applying a vendor patch; a credential-driven intrusion is resolved by invalidating what the attacker stole and making it harder to pivot from one compromised secret to many. That's precisely the shape of Apify's own remediation list: rotate secrets, hash tokens so they can't be read back out even by someone with internal access, restrict network paths between services, and strengthen authentication so a single stolen credential doesn't unlock everything downstream of it. It also means the usual advice to "wait for a patch" doesn't apply here — the actionable response for customers is entirely about their own credentials, tokens, and connected accounts, not about updating any software.

Common questions

For background on why reused credentials and OAuth over-permissioning make incidents like this spread further than the initial breach, see our explainer on how attackers exploit stolen session and account credentials. If you're deciding how to store the rotated Apify API key and other secrets going forward, our comparison of password manager options covers how to manage API tokens alongside passwords. And if your notification mentioned exposed personal data, our guide on freezing your credit after a data breach walks through the follow-up steps.

Frequently asked questions

Is Apify safe to use after this breach?

Apify says it has closed the access path used in the attack, hashed all stored API tokens, added network restrictions, and strengthened internal authentication. There's no indication of ongoing attacker access since November 11, 2026, but customers should still rotate credentials and review connected integrations as a precaution while Apify's post-mortem is pending.

What caused the Apify data breach?

Apify says an unauthorized third party used a compromised credential to access parts of its internal systems between November 9 and 10, 2026, and then used that access to obtain additional credentials and reach related accounts. There is no associated software vulnerability or CVE; this was a credential-compromise intrusion, not an exploited flaw.

What data was exposed in the Apify breach?

Apify's own status-page updates confirm only that some customers' personal information or stored data was accessed. A breach-notification email shared by one affected customer and reported by SecurityWeek listed names, emails, billing details, IP addresses, salted password hashes, 2FA recovery codes, phone numbers, OAuth tokens for Google/GitHub sign-in, and limited card data (cardholder name, expiration date, last four digits only — not full numbers or CVVs). That itemized list is attributed to the customer notification as covered by SecurityWeek, not to Apify's own public statement.

Should I reconnect my GitHub integration on Apify?

Yes, if you build Actors from private GitHub repositories. Apify revoked some OAuth connections, including GitHub, as a precaution during the incident. Go to your integration settings, choose "Change provider," and re-authorize through GitHub rather than assuming the previous connection is still valid.

Do I need to change my Apify password?

If you received a breach notification from Apify, yes — rotate your account password and any API tokens immediately, and treat previously issued 2FA recovery codes as potentially exposed. Even without a direct notification, rotating tokens that existed before Apify's switch to hashed-only token storage removes any residual risk.

What does setting storage access to "Restricted" do?

Apify's default general resource access can let anyone who knows a storage's ID read its contents. Setting it to "Restricted" in Console under Settings → General resource access limits access to your account and anyone you explicitly grant, which Apify is now recommending to all users following the incident.

Sources

More on Apify →Apifydata breachsecurity incidentcredential compromiseweb scraping
Sana Qureshi
Written bySana Qureshi

Sana Qureshi runs the security and privacy desk. She reports on actively exploited vulnerabilities, vendor patches and data breaches, and covers the password managers, VPNs and authentication tools readers use to protect themselves. Her alerts cite vendor advisories, CISA and the CVE record directly.

More from Security & Privacy

See all