Join our Newsletter — 33% off our NHI Course
Home› NHI Breaches› Apify Breach 2026: How One Compromised Credential Led…
Breach analysis Incident: 11 Nov 2026

Apify Breach 2026: How One Compromised Credential Led to Rotated API Tokens, Revoked OAuth Connections and Exposed Customer Data

PDF for newsletter subscribers · free to read online

← All NHI breaches
By NHI Mgmt Group, under the editorial direction of Lalit Choda Updated 20 November 2026 12 min read
On this page

On 11 November 2026, the web scraping and automation platform Apify posted a security notice saying that "an unauthorized third party gained access to parts of Apify's internal systems" on 9 and 10 November 2026. Apify detected the activity at 12:13 UTC on 11 November and cut off the access. In an update the next day, Apify said the attacker "used a compromised credential to obtain a number of account credentials" and reached related accounts. Apify responded by invalidating internal credentials, rotating customer API tokens and proxy passwords, rotating critical internal secrets, resetting all active sessions and revoking some third-party OAuth connections, including GitHub. Affected customers were told to reset two-factor authentication. A customer notice reported by SecurityWeek lists exposed data including 2FA recovery codes and authentication tokens tied to Google and GitHub sign-ins. Apify says it has identified the most likely root cause but has not yet published it, and it has not said whose credential was compromised or how many customers were affected.

Key takeaways

  • Apify says an unauthorised third party accessed parts of its internal systems on 9 and 10 November 2026; it detected the intrusion on 11 November and posted its first public notice that evening.
  • According to Apify, a single compromised credential was used to obtain "a number of account credentials". Apify has not said whether that first credential belonged to a person or a machine.
  • Apify rotated customer API tokens and Apify Proxy passwords, rotated internal secrets, reset all sessions and revoked OAuth connections including GitHub, while saying it had not found those connections to be an attack vector.
  • A customer notice that an Apify customer shared on Reddit, as reported by SecurityWeek, lists names, billing details, salted password hashes, 2FA recovery codes and Google and GitHub sign-in tokens among the exposed data. Apify has not disclosed how many accounts were affected.
  • The identity lesson: once one credential opens a path to stored credentials, every API token, proxy password, OAuth grant and recovery code held by the platform becomes a rotation job for its customers.

At a glance

OrganisationApify (Czech web scraping and browser automation platform) and its customers
WhenUnauthorised access 9 to 10 November 2026; detected 11 November 2026 at 12:13 UTC; first public notice 11 November 2026; marked resolved 20 November 2026
AttackerNot identified. Apify has not attributed the intrusion
Entry pointA compromised credential, according to Apify; whose credential and how it was compromised have not been disclosed
Identities abusedThe initial compromised credential; customer account credentials obtained with it; Apify API tokens and Apify Proxy passwords (rotated); internal secrets (rotated); sessions (reset); third-party OAuth connections including GitHub (revoked as a precaution)
ImpactPersonal information or stored data of some customers accessed, according to Apify; 2FA codes possibly exposed; number of affected customers not disclosed
CategoryNHI. Incident class: Confirmed NHI breach (stolen credential used to obtain stored account credentials; tokens and secrets rotated)

What happened

Apify runs a cloud platform where developers build and run "Actors", programs that scrape websites and automate browsers. Customers drive the platform through API tokens, connect it to source code through GitHub integrations, and route traffic through Apify Proxy using proxy passwords. BleepingComputer reports that Apify says it has 80,000 customers, while SecurityWeek gives the figure as 91,000. Either way, the platform holds a large number of machine credentials that customers embed in scripts, pipelines and AI tools.

Apify's status page notice, first posted at 20:46 CET on 11 November 2026, says that "an unauthorized third party gained access to parts of Apify's internal systems" on 9 and 10 November. Apify detected the activity at 12:13 UTC on 11 November, cut off the access and began an investigation. In its update on 12 November, Apify said the attacker "used a compromised credential to obtain a number of account credentials" and used them to access related accounts. It said that access path was closed and that affected users were being notified individually. Later that day Apify said it had emailed customers whose personal information or stored data had been accessed.

BleepingComputer, which saw sample notices, reports that Apify invalidated internal credentials, rotated API tokens and critical internal secrets, reset all active sessions and revoked some third-party OAuth connections, including GitHub, "out of an abundance of caution". One notice told customers their Apify API tokens and Apify Proxy passwords had been rotated. Another told them to reset two-factor authentication in the Apify Console because 2FA codes may have been exposed, while saying passwords remained safe. SecurityWeek describes a customer notice shared on Reddit by an Apify customer, listing names, email addresses, billing details, account activity, IP addresses, salted password hashes, 2FA recovery codes, phone numbers and authentication tokens tied to Google and GitHub sign-ins, plus partial card details. Full card numbers and security codes were not exposed, according to that notice.

On 13 November Apify advised users to set "General resource access" to Restricted, warning that while storage is open anyone who knows a storage ID can read it. It added: "Be cautious of phishing. Apify will never ask for your password, 2FA codes, or payment details." On 16 November Apify said it had "identified the most likely root cause of this incident", had hardened the attack path and had found no evidence of further unauthorised access. It also changed how tokens are handled: "API tokens are now shown only once, when you create them." On 20 November Apify marked the incident resolved and published a preliminary report on its trust centre, saying a full report will follow. That report could not be read for this page, so its contents are not reflected here. Apify has not disclosed whose credential was compromised, whether it was a human or a machine credential, or how many customers were affected.

Timeline

DateEvent
9 November 2026Unauthorised access to parts of Apify's internal systems begins, according to Apify.
10 November 2026Unauthorised access continues; Apify says the intrusion ran on 9 and 10 November.
11 November 2026Apify detects the activity at 12:13 UTC, cuts off access, revokes some OAuth connections including GitHub and posts its first public notice at 20:46 CET.
12 November 2026Apify says a compromised credential was used to obtain account credentials; affected customers are emailed; BleepingComputer reports token rotation and 2FA reset notices.
13 November 2026Apify advises restricting storage access and warns of phishing; SecurityWeek reports a customer notice listing exposed data.
16 November 2026Apify says it has identified the most likely root cause, hardened the attack path and changed API tokens to be shown only once at creation.
20 November 2026Apify marks the incident resolved and publishes a preliminary report; a full report is to follow.

How it happened: the identity attack path

  1. A credential is compromised. Apify says the attacker used "a compromised credential". It has not said whose it was, what it unlocked or how it was obtained, and this page does not guess.
  2. Internal systems are reached. With that credential the attacker gained access to parts of Apify's internal systems on 9 and 10 November 2026.
  3. Stored account credentials are harvested. From inside, the attacker obtained "a number of account credentials". The customer notice reported by SecurityWeek lists salted password hashes, 2FA recovery codes and Google and GitHub sign-in tokens among the exposed data.
  4. Related accounts are accessed. Apify says the attacker used the harvested credentials to access related accounts, exposing some customers' personal information or stored data.
  5. Detection and mass rotation. Apify detected the activity on 11 November, closed the path, invalidated internal credentials, rotated customer API tokens, proxy passwords and internal secrets, reset all sessions and revoked OAuth connections, pushing re-connection and 2FA resets out to customers.

Impact

  • Confirmed by Apify: unauthorised access to parts of its internal systems; a compromised credential used to obtain account credentials and access related accounts; personal information or stored data of some customers accessed, with those customers notified by email.
  • Confirmed response: customer API tokens and Apify Proxy passwords rotated, internal secrets rotated, all sessions reset, some OAuth connections including GitHub revoked, and customers told to reset 2FA and reconnect GitHub if they build Actors from private repositories.
  • Reported, not verified: a customer notice cited by SecurityWeek lists names, emails, billing details, account activity, IP addresses, salted password hashes, 2FA recovery codes, phone numbers, Google and GitHub sign-in tokens and partial card details. SecurityWeek attributes the list to a notice an Apify customer shared on Reddit.
  • Not disclosed: the number of affected customers, the root cause, whose credential was compromised and who the attacker was. Apify said it found no evidence that the revoked OAuth connections were an attack vector.
  • Potential downstream risk: any customer script, pipeline or AI tool that held an old Apify token or proxy password would break until updated, and any reused password or unrotated third-party token could still be at risk. Apify also warned of follow-on phishing.

What this means for NHI and AI agent security

Apify is on this list because the breach turned on credentials and the platform's response was almost entirely about machine identities. A single compromised credential gave access to stored account credentials, and the clean-up meant rotating every customer API token and proxy password, rotating internal secrets, resetting sessions and revoking OAuth grants to GitHub and other services. Apify's own change after the incident, showing API tokens only once at creation, points to a familiar weakness: a long-lived token that can be viewed again later is a token that can be read by anyone who reaches the account or the system behind it.

The wider lesson is the same one seen at Railway and OpenCode Console: platforms that developers and AI tools plug into hold large stores of customer credentials, and a breach of the platform becomes a rotation exercise for every customer. Scraping and automation platforms are now common data sources for AI agents, so their API tokens often sit in agent configurations and pipelines that nobody inventoried. Treat each such token as a non-human identity with an owner, a scope and a rotation plan, as set out in our API Key Management Guide and Non-Human Identity (NHI) Guide.

The OAuth side matters too. Apify revoked GitHub connections as a precaution even though it found no evidence they were abused. That is the right instinct: an OAuth grant from a breached platform into your source code is a standing path into your organisation, and revoking and re-granting it is cheap compared with finding out later that it was used. Our OAuth Token Theft Incident Playbook covers how to work through that.

Recommendations

  • Replace every Apify token and proxy password you hold. Find each place an Apify API token or proxy password is stored, including CI variables, agent and MCP configurations and scheduled jobs, and replace it with the rotated value. Store new tokens in a secrets manager, since Apify now shows them only once. See our API Key Management Guide.
  • Review and re-grant OAuth connections deliberately. Reconnect GitHub only where it is needed, with the narrowest repository access, and check GitHub and Google for any Apify grants you no longer use. See How to audit GitHub Apps and OAuth apps in an organisation.
  • Reset 2FA and remove password reuse. Reset two-factor authentication in the Apify Console as Apify advised, regenerate recovery codes and change any password reused elsewhere.
  • Check what old tokens could reach and what they did. Review Apify account activity and any downstream systems the tokens touched for use between 9 and 11 November 2026. See How to work out what a leaked API key could access and what it actually did.
  • Restrict storage and data access by default. Set general resource access to Restricted so that storage cannot be read by anyone who knows its ID, and avoid putting secrets in scraped datasets or key-value stores.
  • Plan for third-party platform breaches. Keep an inventory of the credentials you have given each platform, so that mass rotation is a routine runbook rather than a scramble. See our Secret Rotation vs Secret Elimination guide.

Frequently asked questions

What happened in the Apify breach?

Apify says an unauthorised third party accessed parts of its internal systems on 9 and 10 November 2026, using a compromised credential to obtain a number of account credentials and access related accounts. Apify detected the intrusion on 11 November, cut off access, rotated customer API tokens, proxy passwords and internal secrets, reset sessions and revoked some OAuth connections, including GitHub. Some customers' personal information or stored data was accessed.

Do I need to do anything if I use Apify?

Yes. Update any script or pipeline that uses an old Apify API token or proxy password, reset two-factor authentication in the Apify Console, reconnect GitHub if you build Actors from private repositories, set general resource access to Restricted, change any reused password and watch for phishing. Apify says it will never ask for your password, 2FA codes or payment details.

How did the attacker get into Apify?

Apify says the attacker used a compromised credential. It has not disclosed whose credential it was, whether it belonged to a person or a system, or how it was obtained. On 16 November Apify said it had identified the most likely root cause and would publish details in its incident report; a preliminary report followed on 20 November, with a full report still to come.

Railway breach 2026 · OpenCode Console breach 2026 · API Key Management Guide · OAuth Token Theft Incident Playbook · Non-Human Identity (NHI) Guide

How NHI Mgmt Group can help

When a platform you depend on is breached, your team inherits the work: finding every token and OAuth grant you gave it, rotating quickly and checking whether anything was misused. We help organisations inventory the credentials held by third-party platforms, scope and own each one, and build the playbooks that make mass rotation routine. See our NHI and AI agent security training.

References

PDF for subscribers

Take this breach analysis with you

The full analysis in one PDF: timeline, attack path, impact and recommendations, with links to the sources and related pages. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick both boxes to continue.