> ## Documentation Index
> Fetch the complete documentation index at: https://docs.mcpjam.com/llms.txt
> Use this file to discover all available pages before exploring further.

# MCPJam MCP server

> Connect your agent or MCP client to your MCPJam account: projects, saved servers, and evals

Connect your agent to your MCPJam account.

* **Access:** Projects, saved servers, and evals.
* **Client support:** Remote MCP over Streamable HTTP with OAuth sign-in.
* **Setup:** No MCPJam installation or API key required.

## Connect your client

* **Server URL:** `https://mcp.mcpjam.com/mcp`
* **Client differences:** Setup options and approvals vary by client, plan, and workspace.

> [tab] Claude Code
>
> Add the remote server:
>
> ```bash theme={"theme":"css-variables"}
> claude mcp add --transport http mcpjam https://mcp.mcpjam.com/mcp
> ```
>
> Then run `/mcp`, choose **mcpjam**, and sign in with your MCPJam account.
>
> See [Claude Code's MCP guide](https://code.claude.com/docs/en/mcp).

> [tab] Codex
>
> Add the remote server and complete OAuth sign-in in your browser:
>
> ```bash theme={"theme":"css-variables"}
> codex mcp add mcpjam --url https://mcp.mcpjam.com/mcp
> ```
>
> If sign-in does not start during setup, run:
>
> ```bash theme={"theme":"css-variables"}
> codex mcp login mcpjam
> ```
>
> Codex stores MCP server configuration in `~/.codex/config.toml`. See
> [Codex's MCP guide](https://developers.openai.com/codex/mcp).

> [tab] Cursor
>
> Add this entry to `.cursor/mcp.json` in your project, or
> `~/.cursor/mcp.json` for all projects. If the file already exists, merge
> the entry into its `mcpServers` object.
>
> ```json theme={"theme":"css-variables"}
> {
>   "mcpServers": {
>     "mcpjam": {
>       "url": "https://mcp.mcpjam.com/mcp"
>     }
>   }
> }
> ```
>
> Open Cursor's MCP settings and complete the OAuth sign-in for **mcpjam**.
> Enable its tools for your agent. See
> [Cursor's MCP guide](https://cursor.com/docs/context/mcp).

> [tab] VS Code
>
> In VS Code with GitHub Copilot:
>
> 1. Open the Command Palette and run **MCP: Add Server**.
> 2. Choose **HTTP** and enter `https://mcp.mcpjam.com/mcp`.
> 3. Enter **mcpjam** as the server ID and choose where to save the configuration.
> 4. Review the server trust prompt and complete the OAuth sign-in with your
>    MCPJam account when prompted.
> 5. Use Copilot's agent chat and enable the MCPJam tools if they are not
>    already available.
>
> See [VS Code's MCP setup guide](https://code.visualstudio.com/docs/copilot/customization/mcp-servers).

> [tab] Claude
>
> In Claude on the web or desktop:
>
> 1. Open [**Customize → Connectors**](https://claude.ai/customize/connectors)
>    and add a custom connector.
> 2. Name it **MCPJam** and enter `https://mcp.mcpjam.com/mcp`.
> 3. Click **Connect** and sign in with your MCPJam account. The connector shows
>    **Connected** under **Your connectors** when sign-in finishes.
> 4. In a chat, turn it on from **+ → Connectors**.
>
> On Team and Enterprise plans, an owner may need to add the connector for the
> organization first. See Anthropic's
> [connector setup guide](https://claude.com/docs/connectors/custom/add-unlisted#add-a-connector-by-url).

> [tab] ChatGPT
>
> Custom MCP server access depends on your ChatGPT plan and workspace
> permissions. A workspace admin may need to enable or publish the connection.
>
> 1. Open **Plugins**, select **+**, then **Add custom MCP server**. In the
>    legacy Apps experience, use **Settings → Apps → Create** instead.
> 2. Name it **MCPJam** and enter `https://mcp.mcpjam.com/mcp` as the server URL.
> 3. Choose **OAuth**, review the access warning, and sign in with your MCPJam
>    account when prompted.
> 4. Review the discovered tools, finish creating the connection, and select
>    MCPJam in a chat.
>
> If the option is missing, check OpenAI's
> [custom MCP server guide](https://help.openai.com/en/articles/12584461-developer-mode-and-mcp-apps-in-chatgpt-beta)
> for current plan requirements and admin controls.

### Other MCP clients

Add a remote MCP server named **MCPJam** with the URL
`https://mcp.mcpjam.com/mcp`, select **Streamable HTTP** if your client asks for
a transport, and complete OAuth sign-in in your browser. Clients that only
support local stdio servers cannot connect directly to this hosted endpoint.

## Before you start

* **An MCPJam account.** Sign up at [app.mcpjam.com](https://app.mcpjam.com).
  The free plan works.
* **A project.** Your agent can list your projects and create a new one, and every
  project tool falls back to your most recently updated project when you don't
  name one.
* **MCP servers to work with** (optional). Servers you already saved in the
  MCPJam app are available right away. Your agent can also connect a new server by
  URL; when that server needs OAuth, it gives you a sign-in link to open.
* **Plan and feature access.** Some features depend on your plan, your project
  role, or features enabled for your account. Ask your agent what you can do in a
  project. It reads that from your account, and a tool you can't use is
  refused with a clear error.

| | |
| - | - |
| Server URL | `https://mcp.mcpjam.com/mcp` |
| Transport | Streamable HTTP |
| Sign-in | OAuth with your MCPJam account (`login.mcpjam.com`) |
| Acts as | You. Every call uses your own MCPJam identity and your own project access |

> [!NOTE]
>
> This server connects your agent to your MCPJam account. To inspect MCP
> servers locally through an agent instead, see the [CLI MCP server](https://docs.mcpjam.com/cli/mcp-server.md).

## Verify the connection

Ask your agent:

> Use MCPJam to show which account I'm signed in with and list my projects.

Check that it calls `get_me` and `list_projects` and returns the account and
projects you expect. This checks the connection without changing your data
or spending MCPJam credits.

## Sign-in and account access

The server is an OAuth 2.0 protected resource. Your client discovers MCPJam's
sign-in page from the server, you sign in with your MCPJam account, and the client
receives a token for this server. Any request without a valid token is refused
with `401`, which is what starts sign-in.

* **Your access, no more.** Each tool call is forwarded to MCPJam's API with
  your token, so you see exactly the organizations and projects your account
  can see, and writes need the same role they need in the app.
* **Nothing stored by the server.** The server is stateless. It keeps no
  session and no copy of your token between requests.
* **Credentials you save stay write-only.** Server headers, environment values
  and client secrets you save are encrypted at rest. No tool returns a stored
  secret's value.
* **Disconnecting.** Remove or disable MCPJam in your client's MCP server,
  connector, or app settings to stop that client from calling MCPJam.

## Tool permissions

Each tool declares what kind of action it is:

* **Read-only** tools only read your account: listing projects, reading a run's
  results, fetching a trace.
* Tools that **create** something new, such as a project, an eval suite or a
  run, are not read-only.
* Tools that **change, replace, cancel, revoke or delete** something that
  already exists are marked destructive.
  So are `call_server_tool` and `send_chat_message`, because they run tools on
  your own MCP servers, and those tools can do anything.

Your client controls whether it asks for approval before running a tool.
Review its tool permissions before enabling writes or credit-spending
actions. Do not assume every client will prompt before running them.

## Costs

Most tools are free to call. These spend your organization's MCPJam credits,
or the model-provider keys your organization configured, and their
descriptions say so:

| Tool | Spends on |
| - | - |
| `run_eval_suite`, `run_eval_case` | Model calls for every iteration of every case |
| `request_eval_run_judge`, `backtest_eval_run_judge` | LLM-as-judge grading |
| `send_chat_message` | Each Playground message |

These are included with MCPJam and are not charged to you, but they count
against per-minute and daily usage limits: `generate_eval_cases` and
`import_eval_cases`. A tool that hits a limit is refused with `RATE_LIMITED`
and the time it can be called again.
Adding credits does not lift these limits.

## Examples

**See where things stand**

> Which MCPJam projects do I have, and which servers are in the most recent one?

Your agent uses `list_projects`, then `list_project_servers`.

**Check a server**

> Diagnose the "billing" server in my Payments project. It stopped connecting.

Your agent uses `diagnose_server`, which probes the URL, connects, and reports
what failed.

**Run and read evals**

> Run the "checkout flows" eval suite in Payments and tell me which cases fail.

Your agent uses `run_eval_suite`, then `get_eval_run` until the run finishes,
then `get_eval_run_steps` for a failing case. The run spends credits; check
your client's approval settings before requesting it.

**Did a change break anything?**

> Compare the latest "checkout flows" run with the one before it.

Your agent uses `compare_eval_run`, which classifies each case as regressed,
fixed, new, removed, changed, or unchanged.

**Add a server**

> Connect [https://example.com/mcp](https://example.com/mcp) to my Payments project.

Your agent uses `connect_project_server`. If the server needs OAuth, it gives
you a sign-in link to open, then calls `get_project_server_connection_status`
until the connection is ready.

## Troubleshooting

**Your client asks you to sign in again.** Your MCPJam sign-in expired or was
revoked. Reconnect MCPJam from your client's MCP server, connector, or app
settings and complete OAuth sign-in again.

**Your client cannot add the server or start sign-in.** Confirm it supports
Streamable HTTP and OAuth for remote MCP servers, and use the full
`https://mcp.mcpjam.com/mcp` URL. Check whether your client plan or workspace
policy restricts custom servers. Do not configure this endpoint as a local
stdio command or use the MCPJam CLI's stdio configuration for it.

**You see a different account's projects.** You signed in with another
MCPJam account. Disconnect MCPJam in your client, connect again, and choose the account
you meant on the MCPJam sign-in page. `get_me` shows which account your client is
using.

**`NOT_FOUND` for a project, server, suite or run.** The name or id did not
match anything your account can see. Project tools default to your most
recently updated project, so name the project when you have several.

**`403 FORBIDDEN`.** Your role or plan does not allow that action, or the
feature is not enabled for your account. A refusal for a feature carries
`details.code: "FEATURE_UNAVAILABLE"` and names the feature.

**`RATE_LIMITED`.** A usage limit was reached. The error says when the tool can
be called again.

**`TARGET_REQUIRED` from `run_eval_suite`.** The suite is attached to several
environments or clients. The error lists them; tell your agent which one to run.

**Anything else.** Email [support@mcpjam.com](mailto:support@mcpjam.com) with
the request id from the error message when there is one.

## Privacy and terms

The MCP server processes your MCPJam data under MCPJam's
[Privacy Policy](https://www.mcpjam.com/privacy-policy) and
[Terms of Use](https://www.mcpjam.com/terms-of-use).

* Tools that run models (evals, judges, Playground messages, and generation
  tools) send the content they work on to the model
  providers MCPJam uses.
* `call_server_tool` and `send_chat_message` send their inputs to your own MCP
  servers.
* `send_feedback` sends the text you give it to the MCPJam team, where it is
  kept for 180 days.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.