GitHub’s Octoverse 2026 counted 34.3 billion commits, and over 40% of pull requests now start with an AI agent. Read to learn what the State of the Octoverse 2026 Report reveals about AI, citizen developers, and our new security reality.
GitHub’s Octoverse 2026 counted 34.3 billion commits, and over 40% of pull requests now start with an AI agent. Read to learn what the State of the Octoverse 2026 Report reveals about AI, citizen developers, and our new security reality.
Rotating 1,000+ service credentials across 150+ applications with zero outages took one lesson: rotation is the last step. Here’s the six-stage SCALER method, from inventory to rotation, and why a revert path made it safer.
GhostAction pushed 72,026 commits to 22,964 repos in nine days, now targeting CI runners and developer machines with a launcher copied from PolinRider.
GitGuardian researchers reached 44 active Kubernetes clusters through leaked secrets, and 30% had been exposed for more than two years. Here’s how the OWASP Kubernetes Top 10 explains these Kubernetes security risks, and why K03, Secrets Management Failures, turns one leak into a chain.
On Oct. 21, 2026, PHP fixed a vulnerability named php2shell by Calif. Pre-auth RCE under default configurations, confirmed on common CMS, webmail, e-commerce, file sharing, and database administration applications. A PHP server also stores and uses credentials, so this is a secret theft vector.
Railway's security incident may have exposed decrypted customer environment variables. Pull and scan yours with GitGuardian to know what to rotate first.
Developer laptops hold about 150 secrets each, and roughly 28% of secret incidents happen outside code repositories. Learn what credential layer security is and the six controls that protect every surface where credentials live
Researchers pulled real, leaked API keys out of three LLMs using only API output. Here is how AI agents carry secrets from developer laptops into model context, and how to stop them before the model sees them.
The first thing GitGuardian ever shipped was a free warning. Now we’re making sure it gets heard.
Six AI agent security best practices for teams running coding agents and MCP servers: limit what a compromised agent can reach and catch credential misuse.
An AI agent runs with your credentials and acts before you can stop it. Nine incidents show what that costs. Six controls limit the damage: sandbox the agent, scope its credentials, lock its configuration, log every call, scan the workspace first, and deny by default.
AI coding assistants are unpredictable by design. Learn how AI hooks add deterministic controls and how GitGuardian ggshield blocks secrets before agents can use them.
GitGuardian is the code security platform for
the DevOps generation.
With automated secrets detection and
remediation, our platform enables Dev, Sec, and Ops to advance together
towards the Secure Software Development Lifecycle.
Subscribe to our newsletter to receive the latest content and updates from GitGuardian.
