What does Token Forge Cloud do?

Token Forge Cloud provides managed APIs for Qwen, DeepSeek, GLM, and Seedance, with additional video and speech models for teams building production AI products. Proven workloads can also move into a Neo Cloud marketplace product path.

Who is Token Forge Cloud built for?

It is built for product and enterprise teams that need reliable language, reasoning, coding, agent, and video model access without operating every provider stack themselves.

Can teams start with APIs before buying through Neo Cloud?

Yes. Teams can begin with managed API access for Qwen, DeepSeek, GLM, Seedance, and additional media models, then move predictable production traffic into a Neo Cloud marketplace product path.

Which AI workflows does Token Forge Cloud support?

Token Forge Cloud supports assistants, reasoning, coding, agents, text-to-video, image-to-video, asynchronous jobs, model routing, and speech generation.

How does Token Forge Cloud handle security and data control?

Token Forge Cloud supports role policies for who can call which models and workflows, controlled access for prompts and application data, and audit-ready request, generation, and routing telemetry.

How to Add an MCP Server to Claude Code

To add a remote MCP server, run claude mcp add --transport http <name> <url> in a terminal. Replace the placeholders with the server name and URL supplied by its provider. Then start Claude Code and run /mcp to check the connection and approval status. Claude Code must be installed and authenticated first. (Claude Code quickstart)

Add a remote HTTP server

Get the server details. Use the URL and any authentication instructions supplied by the server provider. Requirements vary by server.
Open a terminal in the project where you want to register the server. Run the command in the terminal, not at the Claude Code prompt:
```
claude mcp add --transport http <server-name> <server-url>
```
For example, Anthropic’s quickstart registers its documentation server like this:
```
claude mcp add --transport http claude-code-docs https://code.claude.com/docs/mcp
```
claude-code-docs is a name you choose; the URL identifies the server. Claude Code prints a registration confirmation and a File modified: line. That confirms the entry was saved, not that the server is connected. (Claude Code quickstart)
Check the connection. Start a Claude Code session and enter /mcp. Review the server’s connection status and whether it needs approval for this project. If the server requires sign-in, follow that provider’s authentication steps. (Claude Code configuration debugging)

Add a local server that runs as a process

Claude Code can connect to MCP servers that run as local processes. For a stdio server, follow the server provider’s Claude Code-specific registration instructions and Anthropic’s “Add a local server” guidance. Don’t substitute the HTTP command above: the local server’s launch command and arguments are specific to that server.

After registration, start Claude Code and use /mcp to check its connection status.

Choose where the configuration applies

Claude Code offers three scopes. Choose based on who should use the server and where:

Local (the default): Private to you and active in the project where you add it. The server is not automatically active in other projects.
User: Private to you, but available across your projects. Add --scope user to the command; user-scoped server entries are stored in ~/.claude.json.
Project: Shared with the project team through a .mcp.json file at the project root. Add --scope project to the command. Claude Code may ask you to review and approve the project’s server configuration; /mcp shows its approval status. (Claude Code quickstart)

For example, add --scope user or --scope project before the server name:

claude mcp add --transport http --scope user <name> <url>
claude mcp add --transport http --scope project <name> <url>

Use local scope for a private, project-specific setup; user scope for a personal server you want in multiple projects; and project scope when the team should share the server definition. Because project configuration is shared, avoid putting personal secrets in a committed .mcp.json; follow the server’s documented approach for credentials.

If the same server name is defined in multiple scopes with different endpoints, Claude Code reports a conflict in claude mcp list and /mcp. Avoid duplicate names rather than assuming one definition silently takes precedence. (Claude Code MCP reference)

Verify the setup and check common problems

A registration confirmation means the configuration was written. In a Claude Code session, use /mcp to check connection status and project approval. If it is not connected:

Confirm you are in the intended project, especially if you used the default local scope.
Check that the HTTP URL or local launch command matches the server provider’s instructions.
Complete any server-specific sign-in or credential setup.
Review the server and approve its project configuration if prompted. Anthropic advises connecting only to servers you trust; servers that fetch external content can expose you to prompt-injection risk. (Claude Code MCP reference)

Is a production model API required?

For the registration steps above, Claude Code needs to be installed and authenticated, and you need the server’s connection details. The Claude Code instructions do not say that a separate Token Forge Cloud production model API is required to register an MCP server. An MCP server may itself connect to an API, but that is a separate, server-specific integration. The available Token Forge Cloud product information does not establish a Token Forge MCP setup for Claude Code, so this tutorial does not prescribe a Token Forge endpoint, API key, or model name.

Optional: Connect Claude Code to Rider’s built-in MCP server

If you specifically want JetBrains Rider’s built-in MCP server, Rider’s documentation says it is available starting with Rider 2025.2. In Rider, enable the server under Settings | Tools | MCP Server, then use Clients Auto-Configuration for Claude Code if it appears in the client list. Restart Claude Code for the configuration to take effect. This is a Rider-specific route, not the general setup above; Rider’s documentation does not show the generated Claude Code configuration.

Sources

Connect to MCP servers — Claude Code Docs

Connect Claude Code to tools via MCP — Claude Code Docs

Debug your configuration — Claude Code Docs

MCP Server — JetBrains Rider Documentation

Token Forge Cloud