What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
GitHub now uses AI to find generic, unstructured passwords in source code, the kind that have no recognizable token prefix and that pattern matching tends to miss. Those AI-detected password alerts were not new on October 7, 2026, when GitHub announced a purpose-built detection model and moved existing AI-detected alert scans onto it. The newer part is push-time checking: GitHub also announced AI checks inside push protection that aim to catch unstructured credentials before they enter repository history. As of that announcement, the push-protection AI checks were in private preview, so teams should not assume they are generally available.
Two capabilities, two rollout states
The announcement covers two different features, and they are easy to conflate. One finds passwords that already sit in a repository. The other is meant to stop them at the point of a push. The table below separates them.
As an Amazon Associate I earn from qualifying purchases.
| Capability | What it does | Status as of October 7, 2026 | Billing |
|---|---|---|---|
| AI-detected secret alerts (generic alerts) | Scans repository content for unstructured secrets such as passwords and creates secret-scanning alerts | Live. Existing scans were moved to the new purpose-built model. | Included with GitHub Secret Protection (GHSP) and GitHub Advanced Security (GHAS) at no additional charge, per GitHub |
| AI checks in push protection | Checks for unstructured credentials at push time, giving the contributor a chance to remove the secret before it enters history | Private preview. An administrator must enable it, subject to organization or enterprise policies. | Opt-in; consumes GitHub AI Credits |
AI-based secret checks in the Copilot /security-review command |
Adds AI secret checks to the security-review command | Described as forthcoming in private preview; not yet confirmed as available | Opt-in; consumes GitHub AI Credits |
| Remote MCP server scanning from AI coding agents | Scans changes from compatible clients before commit | Documented in GitHub Docs (accessed October 8, 2026) | Not stated in the sources reviewed |
How the new model finds passwords
Older secret scanning relies largely on known token formats: a cloud provider key or an access token has a prefix and a length that a pattern can match. A password has no such shape. GitHub’s October 7, 2026 Changelog entry, “Purpose-built model for leaked secret detection,” describes the new approach this way: “It reads surrounding code to identify likely credentials, including passwords without a recognizable token format, without generating code or prose.”
Two practical points follow. The model judges likely credentials from context, such as a variable named for a database login placed beside a connection string. And because it is a classifier that flags candidates rather than a rule that matches a fixed pattern, its output is a set of likely secrets that someone should review, not a list of confirmed leaks. That is why the alert workflow matters as much as detection.
#1 Best Overall
- POWERFUL SECURITY KEY: The Security Key C NFC is the essential physical passkey for protecting your digital life from phishing attacks. It ensures only you can access your accounts.
- WORKS WITH 1000+ ACCOUNTS: Compatible with Google, Microsoft, and Apple. A single Security Key C NFC secures 100 of your favorite accounts, including email, password managers, and more.
- FAST & CONVENIENT LOGIN: Plug in your Security Key C NFC via USB-C and tap it, or tap it against your phone (NFC) to authenticate. No batteries, no internet connection, and no extra fees required.
- TRUSTED PASSKEY TECHNOLOGY: Uses the latest passkey standards (FIDO2/WebAuthn & FIDO U2F) but does not support One-Time Passwords. For complex needs, check out the YubiKey 5 Series.
- BUILT TO LAST: Made from tough, waterproof, and crush-resistant materials. Manufactured in Sweden and programmed in the USA with the highest security standards.
What the 2024 beta covered, and what changed
GitHub first announced AI password detection in public beta on July 16, 2024. That launch was narrower in three ways. It applied to Git content only and did not cover non-Git content such as issues or pull requests. It produced alerts in a separate tab from regular secret-scanning alerts. And it was not part of push protection. It was backed by the Copilot API and required a GitHub Advanced Security license at the time, but did not require a Copilot license. Those details describe the 2024 launch; current behavior and plan requirements are those in the October 2026 announcement and current GitHub Docs.
The October 2026 change is therefore two steps: the alert scanner moved to a new model, and push-time checking was added as a preview. Coverage of non-Git content is not addressed in the October 2026 announcement, so readers should not assume issues and pull request text are scanned unless GitHub’s current documentation says so.
Triaging generic alerts
AI-detected findings appear in the generic alerts list, not in the standard secret-scanning list. GitHub warns that generic alerts can have higher false-positive rates and may include secrets that exist only in test fixtures. Plan triage accordingly.
Rank #2
- Requires 3 "AAA" batteries (included)
- Unit auto-locks for 30 minutes after 5 consecutive incorrect PINs
GitHub’s documentation (accessed October 8, 2026) sets several limits that affect large repositories and security reviews:
- Generic alerts are capped at 5,000 per repository, counting open and closed alerts.
- Generic patterns show up to the first five detected locations.
- AI-detected secrets show only the first detected location.
- Generic alerts are excluded from Security overview summary views, so organization-wide counts will not include them.
A security team that relies on summary dashboards should therefore run a direct review of the generic alerts list rather than assume a clean dashboard means a clean repository.
What happens during a push
Push protection already blocks supported secrets from command-line pushes. The established behavior and the new AI checks should be kept distinct. The table of operational outcomes below applies to the established blocking behavior for supported secrets.
- The contributor runs a push. Push protection scans the commits being pushed.
- If a supported secret is found, the push is blocked and the command line shows the detected secrets. It displays up to five at a time.
- The contributor either removes the secret from the commit and pushes again, or uses the bypass path that GitHub provides when the secret is a known false positive.
- If a real credential has reached the remote, treat it as exposed. GitHub advises prompt remediation, with revocation of the credential and removal from repository history as examples. Removing the commit is not sufficient on its own, because the value may already have been copied or cached.
If a push scan times out, the push is not left unscanned. GitHub states that it will scan the commits after the push. That is a recovery mechanism, not a guarantee that the secret never reached the remote, so timeouts should be logged and reviewed like any other exposure.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Plans, billing, and availability
Eligibility depends on plan and platform, and the answers differ by feature.
- AI-detected alerts are included with GHSP and GHAS at no additional charge, according to GitHub’s October 7, 2026 announcement.
- AI push protection requires paid GHSP or GHAS coverage for GitHub Team and GitHub Enterprise Cloud customers. Its opt-in checks consume AI Credits.
- GitHub Enterprise Server 3.23 was described as a planned public-preview destination for AI-detected alerts, included with existing GHSP or GHAS purchase. AI push protection and the Copilot security-review command were not part of that Server release.
Billing for AI Credits is generally attributed to the organization that owns the repository. The announcement describes a separate attribution case for user-namespace repositories owned by enterprise-managed users. Organizations can set SKU-level budgets, but GitHub warns that budget alerts alone do not stop usage, so a budget is a notification, not a cap.
Rank #4
- FIDO-ONLY FUNCTIONALITY: Supports FIDO2 (passkeys) and FIDO U2F protocols for passwordless and second-factor authentication. Does not support OTP, TOTP, Smart Card (PIV), or other advanced features - upgrade to YubiKey 5 Series for extended functionality
- SECURE AND CONVENIENT: Passwordless MFA login with the YubiKey Bio authenticator and biometric information using a fingerprint, with a PIN as a fallback. Simply plug in via USB and use your fingerprint to authenticate
- DEVICE & OS COMPATIBILITY: Compatible with Windows, macOS, ChromeOS, and Linux. Works seamlessly with supported services like Google and Microsoft accounts, and major password managers. See the full compatibility list at "Works With YubiKey"
- DURABLE & RELIABLE: Resistant to tampering, water, and crushing. No batteries or network connectivity required, offering dependable authentication without any downtime. Securely manufactured in USA & Sweden
- Yubico Authenticator App - Fingerprint enrollment, passkey management and PIN configuration available via the app app - Upgrade to YubiKey 5 Series to generate one-time-passwords (OTP) via Yubico Authenticator and for advanced compatibility (OATH, PIV)
For list pricing, GitHub announced on March 4, 2025 that GitHub Secret Protection would be available to GitHub Team customers from April 1, 2025, at $19 per month per active committer. That figure dates from the 2025 announcement, so confirm it on GitHub’s pricing page before budgeting.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Practical setup for a security team
- Confirm the plan and product (GHSP or GHAS) that covers each repository before counting on AI-detected alerts.
- If AI push protection is wanted, check with the administrator who controls organization or enterprise policies; the feature has to be enabled by an administrator.
- Set AI Credit budgets for the SKUs in use and treat them as alerts, not limits.
- Review generic alerts directly, since they are excluded from summary views and can include test secrets.
- For AI coding agents, GitHub’s remote MCP server offers a pre-commit scan. Its findings are ephemeral and do not become persisted alerts, so it complements rather than replaces the alert workflow.
Where this leaves the reader
GitHub is adding AI to find passwords that pattern matching cannot see, and the alert side of that work is already live. The push-time side, which would stop a password before it reaches history, is still in private preview as of October 2026. Until a team can confirm that preview access and its plan eligibility, the reliable protections remain the supported-secret blocking in push protection and prompt rotation of anything that has been pushed.
Sources: GitHub Changelog, “Purpose-built model for leaked secret detection,” October 7, 2026; GitHub Changelog announcement of AI-detected password secret scanning, July 16, 2024; GitHub Docs on secret scanning, generic alerts, push protection, and the remote MCP server (accessed October 8, 2026); GitHub Secret Protection pricing announcement, March 4, 2025.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
