DUBLIN, Calif. — July 6, 2026 — Safeguard today announced secret-key authentication and automatic policy sync for the Guard SDK, the embeddable library developers add to their own AI agents and MCP servers to bring policy enforcement directly into their code. Customers can now generate a secret key from their account Settings and use it to authenticate the SDK in minutes — no manual policy wiring, and no service restart required when policy changes.
Previously, teams configuring Guard's protection inside a custom agent or MCP server needed to hand-wire policy details into their integration. With this release, a single secret key is all it takes: once the Guard SDK is configured with the key, it automatically retrieves the customer's current security policy and enforces it in the background. When the policy is updated in Safeguard, every connected instance of the SDK picks up the change automatically — no redeploy, no restart, no follow-up integration work.
For teams that prefer centralized, credential-free access management over a static key, Safeguard is also introducing an OAuth-based login option for the Guard SDK. Teams can authenticate through their existing identity provider and manage access the same way they manage every other connected tool.
"We built this so that adding real security enforcement to an agent or MCP server takes minutes, not a sprint," said Hritik Kumar Sharma, Founder and CEO of Safeguard. "Generate a key, drop it into the SDK, and your policy is live. When you update that policy, every agent enforcing it updates with it — automatically. That's the whole point of policy as a living contract instead of a one-time configuration step."
Secret-key generation is available today from account Settings for all Safeguard customers, and the Guard SDK with automatic policy sync and OAuth login is rolling out now across supported languages and frameworks.
About Safeguard
Safeguard is the software supply chain security platform that fuses 11 scanners, a security-only AI model lineup (Griffin · Eagle · Lion), and reachability-aware reasoning to find what pattern scanners miss — from CVEs to candidate zero-days — and to ship the fix with cited reasoning. The platform is built for engineering teams shipping production software and for the regulators auditing them. Learn more at https://safeguard.sh.