Permit MCP Gateway
Authentication, fine-grained authorization, consent, and audit—so your agents can move fast without turning your SaaS and data into open desert.
Deploy AI agents with confidence.
Real-time visibility, intelligent detection,
and enterprise-grade security.
Platform Overview
See It In Action
Every session starts with a real human. The gateway authenticates users via your existing IdP—SSO, OIDC, OAuth 2.1—and binds every agent action to a verified identity.
01/04
Drop-in proxy for any MCP server. No changes to your existing setup.
Salesforcehttps://your-gateway.agent.security/mcp?upstream_mcp=https://api.salesforce.com/platform/mcp/v1-beta.2/sobject-all
GitHubhttps://your-gateway.agent.security/mcp?upstream_mcp=https://api.githubcopilot.com/mcp/
Slackhttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.slack.com/mcp
Google Drivehttps://your-gateway.agent.security/mcp?upstream_mcp=https://gdrive.googleapis.com/mcp
Jirahttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.atlassian.com/v1/mcp
Confluencehttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.atlassian.com/v1/mcp
HubSpothttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.hubspot.com
Notionhttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.notion.com/mcp
Linearhttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.linear.app/mcp
PostgreSQLhttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.neon.tech/mcp
Stripehttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.stripe.com
Snowflakehttps://your-gateway.agent.security/mcp?upstream_mcp=https://<account>.snowflakecomputing.com/api/v2/mcp
MongoDBhttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.mongodb.com
AWS S3https://your-gateway.agent.security/mcp?upstream_mcp=https://knowledge-mcp.global.api.aws
Figmahttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.figma.com/mcp
Zendeskhttps://your-gateway.agent.security/mcp?upstream_mcp=https://mcp.zendesk.com/mcp
Your agents can reach Notion, Google, Salesforce, ticketing, code, infra—in milliseconds. That's powerful. It's also a new blast radius.
Gartner's take: As agentic AI expands, organizations will need “guardian agents” that observe and enforce policy at runtime to keep AI reliable and secure.
Permit MCP Gateway is the runtime enforcement layer—
Purpose-built for Identity + Delegation + Least Privilege
Six layers of security that sit between your agents and your MCP servers. You configure once; the gateway enforces on every call.
Connect your existing IdP—the gateway handles OAuth 2.1 flows, token exchange, sessions, and refresh automatically. No custom auth code required.
Every tool call checked against OPA policies that are auto-generated for your stack. Supports RBAC, ABAC, and ReBAC—pick your model, we enforce it.
Build white-labeled consent screens with the visual editor. Your branding, your rules, IAM-governed. No frontend work needed.
Auto-provision agent identities from your IdP. Policy-based relationships between humans and agents. Plugs into IGA, PAM, IAM.
One control plane for every MCP decision. Real-time monitoring, anomaly detection, compliance reporting—across all servers.
Full decision chain: user, agent, tool, policy, outcome. Searchable, exportable, SIEM-ready.
No SDK to install. No agents to rewrite. No MCP servers to modify.
Just configure, consent, connect.
Auto-Generated Policies
Point the gateway at your MCP servers. It inspects available tools and generates a contextual authorization policy for each one. You pick the trust level; the gateway writes the OPA policy.
Visual Editor, Your Brand
Build white-labeled consent flows with the visual editor. Decide which tool calls need user approval, what gets shown on the consent screen, and who governs the rules. No frontend work needed.
Same Protocol, New Powers
Give developers and users the proxied MCP endpoint. Auth, authorization, consent, and logging all happen at the gateway. The MCP protocol stays the same, so nothing breaks downstream.
Before any tool is exposed, the gateway requires a single handshake: identify_self. It fingerprints the agent and continuously monitors for drift—so you always know exactly what's connecting.
Prevents: Shared client sessions, reused permissions, and invisible privilege escalation.
Every capability designed to survive the chaos of agentic AI at scale—no gaps, no workarounds.
Humans
Bring your IdP. Authenticate delegators via SSO/OIDC and attribute every action to a verified human.
Agents
Get auto-generated policies you can edit in OPA/Rego, plus a Google Zanzibar-inspired relationship graph that powers ReBAC for real-world delegation and multi-hop access chains.
Your Security Team Won't Hate
Trust slider for tiered trust. Tool picker for explicit allowed tools. Auto-trust heuristics to reduce friction.
Zero Credential Exposure
Gateway handles OAuth flows, stores tokens in a vault, and issues the agent a stand-in token. Zero standing permissions.
By Default
Every decision becomes an audit trail: what happened, when it happened, and why it was allowed or denied.
Drift Detection
Detect prompt drift before it becomes an incident. Interrogate. Fingerprint. Enforce.
Gartner calls them “guardian agents”—AI-native watchers that observe every action and adjust policies on the fly. Permit provides the dynamic, fine-grained control plane and enforcement point they need to act in real time.
Observe every agent action, tool call, and data access in real time
Dynamically update policies on the fly based on risk signals and drift
Fine-grained enforcement using agentic identity: human delegation, context, and intent
For Internal Agents
Allow Cursor, Claude, and internal agents to use MCP safely. Stop incidents on sensitive surfaces—email, CRM, files, ticketing.
For Your Customers
If you expose MCP as a product surface, get multi-tenant, policy-driven authorization with customer-specific consent and trust tiers.
Why Permit
Most “MCP security” products stop at the gateway. Permit's advantage is years building the authorization control plane underneath.
As agentic AI collapses boundaries, the identity stack converges: IGA + PAM + Zero Trust + IAM acting as one system—real-time, dynamic, and fine-grained.
Permit's hybrid architecture decouples control plane from data plane—keep enforcement close to workloads and push policy updates in real time via OPAL.
Start in minutes. No infrastructure to manage. Permit handles the control plane so you can focus on building.
Keep enforcement and sensitive flows in your environment. Full control over your data plane with Permit's PDP deployed alongside your workloads.
Agents won't always go through the gateway. So we're adding connectors that detect, alert, and block when agents touch sensitive surfaces outside.
“For the first time, we see what’s really happening inside our agentic stack. When something deviates, we know it’s real.”
Enterprise Security Leader
We'll make sure they only go where they're allowed.
Or schedule a call with the Permit team →