Failure pattern · 20 documented incidents
Weak or phishable authentication on a privileged path
A password, an SMS code or a help-desk reset is all that stands between an attacker and admin access.
See it happen
Watch how this failure unfolds, then switch to the same situation with the safeguard in place. Click any step to replay it.
Scroll the diagram sideways to see every system.
A phone call was the only barrier to admin.
What goes wrong:
- An attacker calls the help desk, pretending to be an employee.
- The help desk resets the password and the SMS code over the phone.
- The attacker signs in to the admin console.
- A phone call was the only barrier to admin.
With the safeguard:
- An attacker calls the help desk, pretending to be an employee.
- Resets require verified identity, so the help desk checks with the real person.
- Admin sign-in needs a passkey on the employee’s own device, so the attempt fails.
- The real employee reports the attempt.
What it is
A password, an SMS code or a friendly phone call to the help desk is all that stands between an attacker and admin access. Attackers don’t need to break in when they can be let in.
How it happens
- Privileged accounts use passwords or SMS codes that can be phished
- The help desk can reset MFA based on a phone call
- Remote access portals and contractor accounts have weaker login rules
What it causes
- Attackers with admin access to core systems
- Ransomware and data theft across the company
- Customer accounts taken over through internal tools
Warning signs
- MFA resets requested for privileged users
- Push-notification floods (MFA fatigue)
- Admin logins from new devices or locations
How to prevent it
Require phishing-resistant MFA on every privileged path. Use passkeys or hardware keys for admins, vendors and support tools, verify identity out of band before credential or MFA resets, and keep sessions short-lived and device-bound.
- Privileged access requires phishing-resistant MFA, such as passkeys or security keys
- Help-desk resets for privileged accounts need verified identity, not just a convincing call
- Remote access and contractor accounts follow the same login rules as everyone else
Key ideas:Phishing-resistant MFA
Check your system against the full launch checklistReal incidents with this pattern
Documented incidents first. Each links to its public source.
Self-replicating "Shai-Hulud" worm compromises more than 500 npm packages by stealing developer tokens (2025)
CISA reported that a self-replicating worm known as "Shai-Hulud" had compromised more than 500 npm packages.
Safeguard: Keep publishing credentials out of developer environments
GitHub Actions injection leaks Nx's npm token; malicious packages use local AI coding tools to hunt for secrets (2025)
A pull-request title validation workflow used pullrequesttarget, which runs with repository secrets, and echoed the unsanitized PR title, creating an injection point; the workflow's token also had read/write permissions.
Safeguard: Never run untrusted pull-request input with secrets
Stolen OAuth tokens from the Salesloft Drift integration are used to export data from customers' Salesforce instances (2025)
From as early as August 8 to at least August 18, 2025, an actor tracked as UNC6395 used compromised OAuth tokens of the Salesloft Drift third-party application to access customers' Salesforce instances.
Safeguard: Scope and expire third-party integration tokens
Malicious code committed to the Amazon Q Developer VS Code extension via a compromised token (AWS, 2025)
Malicious code was committed to the extension's open-source repository using a compromised GitHub token in a build configuration.
Safeguard: Protect build credentials and require review for releases
"ToolShell" vulnerabilities in on-premises SharePoint are exploited to steal server keys (2025)
Microsoft observed attackers exploiting on-premises SharePoint Server vulnerabilities (CVE-2025-53770 and CVE-2025-53771, related to CVE-2025-49704 and CVE-2025-49706) against internet-facing servers, with attempts as…
Safeguard: Patch internet-facing servers on a short deadline
Default password and an IDOR in McDonald's AI hiring chatbot platform expose data on 64 million applicants (2025)
McHire, the chatbot recruiting platform used by most McDonald's franchisees, lets applicants chat with a bot named Olivia, built by Paradox.ai, that collects personal information.
Safeguard: Remove default and test accounts before production
Call and text records of nearly all AT&T wireless customers taken from a third-party cloud workspace (2024)
Contractor credentials without MFA accessed third-party cloud data warehouse; downloaded 109 million customer call and text records.
Safeguard: Require MFA for contractor and third-party access
Stolen credentials on a remote-access portal lead to the Change Healthcare ransomware outage (2024)
Single-factor remote access portal breached with stolen credentials; ransomware forced nationwide claims gateway disconnection.
Safeguard: Require MFA on all remote-access portals
Password spray on a legacy test tenant exposes Microsoft corporate email to a nation-state actor (2024)
Legacy test tenant without MFA compromised via password spraying; adversary created malicious OAuth apps with mail.read privileges.
Safeguard: Enforce MFA everywhere, including legacy and test tenants
Stolen credential opens Okta's support system, where uploaded HAR files held customer session tokens (2023)
Support portal allowed unredacted HAR file uploads containing session tokens; threat actor hijacked customer administrator accounts.
Safeguard: Redact secrets from customer-uploaded diagnostics
Help-desk social engineering leads to ransomware and a systems shutdown at MGM Resorts (2023)
Social engineering of IT helpdesk reset admin credentials; ransomware deployed directly to ESXi hypervisors shutting down resort operations.
Safeguard: Verify identity out of band for credential resets
Attack on a senior DevOps engineer's computer leads to theft of cloud backups (LastPass, 2022-2023)
LastPass said an attacker exploited vulnerable third-party software to target a senior DevOps engineer.
Safeguard: Separate privileged engineering access from personal environments
Malware steals an engineer's SSO session and exposes CircleCI customer secrets (2023)
Malware on an engineer's laptop stole a valid two-factor-backed SSO session, letting the attacker impersonate the employee and reach a subset of production systems.
Safeguard: Harden employee endpoints and sessions
Phishing impersonating a CI vendor yields GitHub credentials and 130 source-code repositories (Dropbox, 2022)
Phishing emails impersonating CircleCI led an employee to a fake page where GitHub credentials and a one-time code were captured, giving the attacker access to a GitHub organization and 130 code repositories.
Safeguard: Require phishing-resistant authentication for code hosting
MFA fatigue against a contractor gives an attacker access to Uber's internal tools (2022)
An attacker likely bought a contractor's corporate password on the dark web after malware infected the contractor's personal device, then repeatedly sent two-factor approval requests until the contractor accepted one.
Safeguard: Use phishing- and fatigue-resistant MFA
SMS phishing of employees leads to compromise of Twilio customer accounts (2022)
Attackers used SMS phishing to fool some Twilio employees into giving up credentials, then used them to access internal systems and information about a limited number of customer accounts.
Safeguard: Use phishing-resistant MFA for employees
Stolen OAuth tokens from two integrators are used to download private repositories (Heroku and Travis CI, 2022)
GitHub found an attacker abusing stolen OAuth user tokens that had been issued to two third-party integrators, Heroku and Travis CI.
Safeguard: Minimise and protect stored OAuth tokens
Phone spear-phishing gives attackers Twitter's internal support tools; 130 accounts targeted (2020)
A phone spear-phishing attack on a small number of employees gave attackers credentials to internal support tools on July 15, 2020.
Safeguard: Require two-person approval for high-impact account actions
AWS access key in a private GitHub repository leads to 57 million rider and driver records taken (Uber, 2016)
Between October 13 and November 15, 2016, intruders used an AWS access key that had been posted to a private GitHub repository to access Uber's Amazon S3 datastore.
Safeguard: Keep secrets out of repositories
Payment card data for about 40 million accounts taken from Target's U.S. stores (2013)
HVAC vendor portal credentials lacked two-factor authentication; attackers pivoted into internal networks and pushed BlackPOS memory scrapers to 1,800 checkout terminals.
Safeguard: Constrain third-party vendor access
176 security advisories and illustrative scenarios with this pattern
Astrana Health, Inc. discloses a cybersecurity incident involving social engineering (SEC 8-K, September 2026)
Astrana Health, Inc. (the “Company”) recently became aware that its subsidiary Astrana Health Management, Inc. detected unusual activity within its environment.
Nutex Health Inc. discloses a cybersecurity incident with data taken without authorization (SEC 8-K, August 2026)
As disclosed in Item 8.01 of a Current Report on Form 8-K filed with the Securities and Exchange Commission on August 24, 2026 (the “Prior 8-K”), Nutex Health Inc.
AdaptHealth Corp. discloses a cybersecurity incident involving social engineering (SEC 8-K, July 2026)
AdaptHealth Corp. (the “Company”) is investigating a security incident whereby a threat actor gained unauthorized access to Company systems and exfiltrated certain data therefrom.
iRhythm Holdings, Inc. discloses a cybersecurity incident involving social engineering (SEC 8-K, June 2026)
On June 8, 2026, iRhythm Holdings, Inc. (the “Company”) identified unauthorized activity involving data maintained on certain third-party-hosted business applications.
Bitcoin Depot Inc. discloses a cybersecurity incident with data taken without authorization (SEC 8-K, April 2026)
On March 23, 2026, Bitcoin Depot Inc. (the “Company”) discovered that an unauthorized party gained access to certain of its information technology systems.
Coupang, Inc. discloses a cybersecurity incident (SEC 8-K, December 2025)
On November 18, 2025, Coupang Corp. (“Coupang Corp.”), a wholly-owned Korean subsidiary of Coupang, Inc. (Coupang Corp., together with Coupang, Inc.
Zoomcar Holdings, Inc. discloses a cybersecurity incident (SEC 8-K, June 2025)
On June 9, 2025, Zoomcar Holdings, Inc. (the “Company”) identified a cybersecurity incident involving unauthorized access to its information systems.
Coinbase Global, Inc. discloses a cybersecurity incident (SEC 8-K, May 2025)
On May 11, 2025, Coinbase, Inc., a subsidiary of Coinbase Global, Inc. (“Coinbase” or the “Company”), received an email communication from an unknown threat actor claiming to have obtained information about certain…
Dropbox, Inc. discloses a cybersecurity incident (SEC 8-K, May 2024)
On April 24, 2024, Dropbox, Inc. (“ Dropbox ” or “ we ”) became aware of unauthorized access to the Dropbox Sign (formerly HelloSign) production environment.
Software component issue could expose accounts or credentials
A flaw was found in cifs-utils. This flaw allows an unprivileged local user to achieve privilege escalation. When a privileged service mounts a Common Internet File System (CIFS) share using a user-controlled credentials…
Software component issue could grant elevated privileges
The WP Compress – Instant Performance & Speed Optimization plugin for WordPress is vulnerable to Privilege Escalation in all versions up to, and including, 7.25.00 The vulnerability exists because the startcomms…
Software component issue could expose accounts or credentials
The HUSKY – Products Filter for WooCommerce Professional plugin for WordPress is vulnerable to Sensitive Information Exposure in all versions up to, and including, 1.4.5 via the 'woofsettings[woofstat][serveroptions]'…
In the standards
CollectSEA's mapping; not an official determination by the standards bodies.