> ## Documentation Index
> Fetch the complete documentation index at: https://conductorone-paul-querna-mcp-sources-docs.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Set up an MCP source

> Add an MCP source to C1, link it to a C1 application, and configure authentication so C1 can govern access to its tools and resources.

<Note>
  **Activation required.** AI access management must be enabled for your tenant before you can use it. To get started, [contact the C1 support team](mailto:support@c1.ai) for a walkthrough.
</Note>

This page walks through adding an MCP source to C1, linking it to a C1 application, and configuring authentication. Once added, C1 discovers the tools and resources the source exposes. See [Govern tools and toolsets](/product/admin/tools-and-toolsets) and [Govern MCP resources](/product/admin/mcp-resources) for what to do next.

## Before you begin

* AIAM must be enabled for the tenant. See [Enable AI access management](/product/admin/enable-ai-access-management).
* For OAuth-based auth, you'll need a client ID and secret from the downstream service — unless the source's MCP server supports OAuth Dynamic Client Registration (DCR), in which case C1 registers itself automatically and no credentials are required.
* For per-user OAuth, the downstream service must be reachable by C1's hosted callback URL.

## Add an MCP source

C1 walks you through adding an MCP source one question at a time, whether you choose a source from the catalog or provide an external URL:

<Steps>
  <Step>
    In C1, go to **AI > MCP Sources** and click **Add MCP source**.
  </Step>

  <Step>
    Choose a Catalog MCP source from the shortlist or browse all available sources, or add an External MCP source using a custom URL.
  </Step>

  <Step>
    Answer the setup questions as they appear: where the source's MCP server is hosted (public or through a bridge), the server URL, who should be able to use it and when, and what happens to newly discovered tools.
  </Step>

  <Step>
    Confirm authentication. C1 probes the MCP server's OAuth configuration automatically — a self-registering (DCR) server asks nothing further here. Otherwise, choose an auth method; see [Configure authentication](#configure-authentication) for what each one needs.
  </Step>

  <Step>
    Review and submit. C1 adds the source and runs initial discovery.
  </Step>
</Steps>

<Warning>
  **Access profile selection isn't applied yet.** The guided flow lets you choose an access profile during setup, but that selection currently isn't bound to the source — it's recorded but not enforced. Assign an access profile afterward from the source's settings, or through [Govern tools and toolsets](/product/admin/tools-and-toolsets), rather than relying on the in-flow selection.
</Warning>

After setup, the source appears in **AI > MCP Sources**. Its discovered tools are listed on the **Tools** tab. If the source exposes resources, they are listed on the **Resources** tab. You can sort the **Tools** tab by tool name, visibility, classification, state, and last updated.

<Note>
  The source's URL can't be changed once it completes its first successful sync. To point at a different endpoint, add a new MCP source instead.
</Note>

## Configure authentication

C1 supports multiple auth methods for downstream MCP servers. Admins can select any supported method when configuring a source.

| Method                            | Who the downstream sees                                                     | When to use                                                                                                 |
| :-------------------------------- | :-------------------------------------------------------------------------- | :---------------------------------------------------------------------------------------------------------- |
| **Bearer token**                  | A single shared service account                                             | Simple integrations where per-user attribution downstream isn't required                                    |
| **Custom header**                 | A single shared identity via a custom HTTP header (for example, an API key) | Services that authenticate via a non-standard header or API key                                             |
| **Basic auth**                    | A single shared identity via username and password                          | Services that use HTTP Basic authentication                                                                 |
| **Client credentials**            | A single shared OAuth client                                                | Machine-to-machine OAuth where per-user identity isn't needed                                               |
| **OAuth2 — service mode**         | A single shared identity (admin authenticates once)                         | When the downstream requires an OAuth auth code flow but a single shared credential is acceptable           |
| **OAuth2 — per-user passthrough** | Each end user, with their own credentials                                   | When the downstream needs per-user identity (Google Workspace, GitHub, Salesforce, and so on)               |
| **JWT Bearer (RFC 7523)**         | A service identity via signed JWT                                           | Services that support certificate-based or JWT-based auth (for example, Tableau or Google service accounts) |

For per-user OAuth passthrough, C1 vaults each user's downstream tokens and auto-refreshes them so end users don't hit token expiry mid-session.

To configure auth:

<Steps>
  <Step>
    From the MCP source's settings, click **Edit authentication**.
  </Step>

  <Step>
    Select the auth method.
  </Step>

  <Step>
    Enter the required credentials for the selected method:

    * **Bearer token** — paste the token. C1 vaults it.
    * **Custom header** — enter the header name and value.
    * **Basic auth** — choose a credential mode:
      * **Shared (admin authorizes)**: Enter a username and password once. All users connect using the same credentials.
      * **Per-user (each user submits their own)**: No credentials to enter. Each user provides their own username and password when they connect, and MCP requests run under their individual identity.
    * **Client credentials** — enter client ID, client secret, and token URL.
    * **Service mode** — enter client ID, client secret, authorization URL, token URL, and scopes. An admin completes the OAuth flow once; all users share that credential.
    * **Per-user passthrough** — enter client ID, client secret, authorization URL, token URL, and scopes. End users see a Connect prompt the first time their AI client calls a tool from this source.
    * **JWT bearer** — enter the issuer, private key, subject, audience, token URL, and scopes.

    If the source's MCP server supports OAuth Dynamic Client Registration (DCR), you can skip entering a client ID and secret entirely. Toggle on **Use dynamic client registration** — C1 registers itself with the server's authorization server automatically.

    Before adding the source, C1 runs a discovery check against the server's authorization server. If discovery finds an issue that can be relaxed — such as an issuer mismatch or an authorization server on a different host — C1 shows you the finding and asks you to acknowledge it before setup continues. Findings that can't be safely relaxed block setup entirely.
  </Step>

  <Step>
    Click **Save**. C1 makes a test call to validate the credentials.
  </Step>
</Steps>

## Configure source settings

| Setting              | Required?                                                    | What it does                                                                                                                                                                                                                                             |
| :------------------- | :----------------------------------------------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Authentication**   | Required                                                     | See above                                                                                                                                                                                                                                                |
| **Data sensitivity** | Optional                                                     | Metadata tag on the source (low / medium / high). Surfaces in the catalog and audit log; no enforcement                                                                                                                                                  |
| **Tool prefix**      | Optional; required if multiple sources are linked to one app | Prepended to tool names so AI clients can disambiguate (for example, `gh_` vs `gl_` for two Git providers). If you don't set one, C1 uses the prefix the source's MCP server implementation declares; if it doesn't declare one, C1 generates a default. |
| **Tool enablement**  | Optional                                                     | Choose **Require enablement** or **Auto-enable** for this source's tools, independent of your tenant's [default tool approval setting](/product/admin/enable-ai-access-management#require-tool-approval). Leave unset to inherit the tenant default.     |

## What happens after setup

* C1 runs an initial **tool and resource discovery** sweep against the source. Discovered tools appear under the **Tools** tab with state **Pending Review** by default. Discovered resources and URI templates appear on the **Resources** tab with state **Pending**.
* C1 re-runs discovery on a schedule. New tools and resources need review. Tools that disappear are flagged but not auto-deleted. Resources that disappear are marked **Removed** and cannot be read.
* No tool from this source is callable by any end user yet — see [Govern tools and toolsets](/product/admin/tools-and-toolsets) to approve, classify, and bundle them. A resource also needs approval and a user grant before it can be read. See [Govern MCP resources](/product/admin/mcp-resources).