Sylo — The security gateway for AI agents

The security gateway
for AI agents

Sylo sits between your sandboxed agents and the APIs they call. Inject credentials, enforce policies, scan for PII, and audit everything — without giving agents your API keys.

Try Cloud Read the Docs GitHub

import { SyloClient, credential } from "@runplex/sylo";

const session = await sylo.createSandboxToken({
  credentials: [
    credential("anthropic", process.env.ANTHROPIC_API_KEY),
  ],
  policies: { default: "deny" },
});

// Launch sandbox — credentials injected transparently
const sandbox = await Sandbox.create(template, {
  envs: sylo.sandboxEnv(session),
});

Credential Injection

API keys encrypted at rest, injected into requests on the wire. Agents make normal API calls — they never see real credentials.

Policy Enforcement

Control which domains, methods, and paths your agents can access. Default-deny or default-allow. Rate limiting per domain.

PII Scanning

Detect sensitive data in request bodies before they reach external APIs. Redact in-flight or block the request entirely.

Audit Logging

Every proxied request logged with tenant, user, destination, and policy decision. Query by any field for full visibility.

Bypass-Proof

Tunnel mode redirects all outbound TCP through the gateway via iptables. No bypass without root — which sandboxes don't give.

Self-Hosted

One docker compose up on your infrastructure. Caddy for auto-TLS, Postgres for storage, Redis for speed. Or use managed.

How it works

Sylo intercepts outbound requests from sandboxed agents, injects credentials, enforces policies, and forwards to the real API.

Ready to secure your agents?

Get started in under 5 minutes with the managed instance or self-host on your own infrastructure.

Read the docs Self-host