Services

Start a project Chat on WhatsApp
News & Insights

Hackers Now Use AI to Steal Leaked Passwords From GitHub in Five Minutes: Check Your Business Today

A man at a desk in an office using a laptop showing a website, with sunlight across the desk

What happened

Microsoft has warned that attackers are now using AI agents to hunt for leaked passwords and keys in public code, and to break into cloud accounts in minutes. In a report published on 5 November 2026, Microsoft Threat Intelligence said threat actors are using an agentic AI framework called ARTEX to search public GitHub repositories for exposed secrets, including application credentials and API keys for AI services.

In one case Microsoft observed, an AI agent went from finding a leaked secret in GitHub to stealing data from an Azure Key Vault in five minutes. The activity has been running since August 2026. From 28 October, attackers used stolen application credentials to sign in to Microsoft Entra ID accounts across several organisations, and in at least one case a vendor’s own application credential was compromised and then used to reach that vendor’s customers. Once inside, the attackers asked for storage keys, connection strings and app settings, and opened SharePoint, OneDrive and mailbox items.

What it means for a small business

This is not only a big company problem. Any small business whose website or app was built by a developer may have a password, API key or connection string sitting in public code without knowing it. A common shortcut is pasting a key into a config file to get something working quickly. That shortcut is now the fastest way in, because an AI agent can scan thousands of repositories and test what it finds in minutes, work that would take a human attacker hours or days.

Deleting the leaked secret from the code does not fix it. Microsoft is clear on this point: a key that was exposed still works until it is revoked or rotated, even after the code that contained it is gone.

What to do about it

  1. Ask your developer to scan every code repository for secrets, and check any GitHub secret scanning alerts.
  2. Rotate every key or password that was ever exposed, even if it was later deleted from the code.
  3. Never put passwords or keys in code again. Keep them in environment variables or a secrets manager.
  4. Turn on push protection, so a secret cannot be committed by accident.
  5. Where possible, use managed identities instead of long-lived secrets.
  6. Review sign-in logs for your cloud and email accounts for logins you do not recognise.
  7. If you do not have a developer on hand, get one to run this check. It is a small job that prevents a large breach.

Digilistan builds websites and apps and runs hosting and security checks for small businesses in the United Kingdom, the United States and Pakistan. If your site was built by someone else and you have never had its code checked for leaked secrets, that check is the right first step.

A note on what is not yet known: Microsoft has not named the affected organisations. A separate security firm reported related credential abuse at 27 organisations, but it said the link to a specific AI framework was low confidence, so treat that part as unconfirmed.

Start the conversation

Tell us what you are building

Whether you need a website, a chatbot that never sleeps or a workflow that gives your team its evenings back, start with a short conversation. We reply to every serious enquiry personally.

  • Your message goes to the founder, not a call center.
  • We suggest the smallest service that fixes the problem.
  • The first conversation costs nothing but a few minutes.

Prefer to talk first? Message us on WhatsApp and skip the form.

Chat with us