Riptides — Runtime Identity & Security for AI Agents and WorkloadsToggle Menu

Who is Riptides for?

Riptides is built for security, platform, and infrastructure teams securing both AI agents and traditional workloads. Whether your agents make API calls and connect to MCP servers, or your services need secretless credentials and mTLS — Riptides secures those connections at the kernel level with no code changes.

What's the difference between Agentic AI Security and Workload IAM?

Both products share the same kernel-level enforcement engine. Agentic AI Security focuses on per-agent identity, egress control for tool calls, and credential protection for AI agents. Workload IAM focuses on secretless infrastructure, automatic mTLS, and identity federation for traditional services and CI/CD pipelines. Many teams use both.

Which agent frameworks and languages does Riptides work with?

All of them. Riptides operates at the kernel level, below the application layer, so it works with LangChain, CrewAI, AutoGen, OpenAI Agents SDK, MCP-based agents, and any custom agent or service that makes network calls. Any language, any framework — no integration required.

How is Riptides different from a service mesh or AI gateway?

Service meshes and AI gateways operate at the application or network layer — they require sidecars, proxy configuration, or routing changes. Riptides enforces security at the kernel. No sidecars, no proxies, no code changes. Workloads cannot bypass, disable, or route around it.

What makes kernel-level enforcement better?

Kernel-level enforcement means identities are tied directly to running processes, not to network locations or application logic. It's fully independent of both the application code and the underlying network stack. This prevents identity spoofing and blocks unauthorized connections even from compromised processes.

How does secretless credential injection work?

When a workload makes an outbound request, the Riptides kernel module intercepts it, matches the workload's SPIFFE identity to a credential binding, fetches the credential from your secret store (Vault, K8s Secrets, AWS Secrets Manager), and injects it into the request. The credential never enters process memory — it exists only in kernel space for the duration of the request.

Is Riptides compatible with my existing infrastructure?

Yes. Riptides integrates with Kubernetes and container platforms, supports open standards like SPIFFE and mTLS, federates to AWS/GCP/Azure via OIDC, and streams telemetry to your existing observability stack (OpenTelemetry, Datadog, Splunk, Elastic). It can be introduced gradually without requiring a full platform overhaul.

Does Riptides store secrets outside my environment?

No. The Riptides control plane never stores your secrets or credentials. Identity issuance, credential injection, and policy enforcement all happen within your environment. Sensitive data never leaves your trust boundary.