Question 1

How many MCP servers do you support?

Accepted Answer

Two answers, both true. A core set of SaaS apps are C1.ai-supported — formally modeled, risk-classified, conformance-tested against the live API, and covered by our response-time SLAs. A larger set are C1.ai-governed: formally modeled, risk-classified, validated through 37+ structural and semantic checks before each release. Beyond that, you can connect any MCP server you bring — vendor-published, community-built, or one your team built — running anywhere from a hosted endpoint to your cloud or on-prem through our tunnel. We don't publish a single big number. Other catalogs inflate counts by scraping GitHub or aggregating public OpenAPI lists. What we publish is what we support.

Question 2

Why not just use the MCP servers on GitHub?

Accepted Answer

You can. A lot of people are. Those vibe-coded MCP servers are the modern equivalent of downloading random software from a pop-up ad in the '90s. They're often poorly documented, sometimes outright sketchy from a supply chain perspective. When you connect a native MCP server directly to your agent, that agent inherits your full set of entitlements in the target application. There is no separation between your human access and your agent's access. The C1.ai MCP gateway changes that equation: we broker the connection, store the credentials in our vault, and enforce granular policies on what your agent can and cannot do — independent of your standing human access.

Question 3

What if you don't have an integration for the SaaS app I need?

Accepted Answer

Connect any native MCP server — paste a URL and we introspect it. Vendor-published, community-built, or one your team built. The C1.ai gateway authenticates, authorizes, and audits every tool call the same way regardless of where the MCP came from.

Frequently asked questions

Catalog and coverage

Auth modes and credentials

Agents and Service Principals

Deployment and pricing

Still have questions?

Frequently asked questions

Still have questions?

Frequently asked questions

Catalog and coverage

How many MCP servers do you support?

Why not just use the MCP servers on GitHub?

What if you don't have an integration for the SaaS app I need?

Auth modes and credentials

Which MCP servers can I connect, and how does auth work?

Where do API tokens and credentials live?

How does C1.ai work with SAML-protected SaaS apps?

Agents and Service Principals

How is agent access different from human access?

What about enterprise agents that run autonomously?

What approval flows are available?

Deployment and pricing

Is this a separate product? Do we need a new contract?

Do you have a VPC deployment or on-prem version of C1.ai?

Can this save us software licenses?

What about logging, auditing, and compliance?

Still have questions?

Frequently asked questions

Still have questions?