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

# Set up coding agents

> Configure Claude Code, Codex, Gemini CLI, and Deep Agents to route LLM calls through the LLM Gateway.

<Note>
  **Private beta:** The LLM Gateway is in private [beta](/langsmith/release-stages). [Sign up for the waitlist](https://www.langchain.com/langsmith-llm-gateway-waitlist) to get access.
</Note>

Configure coding agents to route LLM calls through the LLM Gateway, giving your organization cost controls, observability, and audit trails on developer LLM usage without changing agent code.

Most coding agents let you override the LLM provider endpoint via environment variables or configuration files. By pointing them at the LLM Gateway instead of the provider directly, all LLM calls flow through the gateway. The gateway authenticates the caller, resolves the actual provider key from workspace secrets, enforces policies, and traces the call. No agent code changes required.

The gateway supports Anthropic, AWS Bedrock, Baseten, Fireworks, Google Gemini, Google Vertex AI, and OpenAI. For the full list and required secrets, see [Supported providers](/langsmith/llm-gateway#supported-providers).

## Prerequisites

* Your [Organization admin](/langsmith/rbac#organization-admin) has completed [Admin setup](/langsmith/llm-gateway-admin-setup).
* You have a workspace-scoped [LangSmith API key](/langsmith/create-account-api-key) with `gateway:invoke` and `workspaces:read` [permissions](/langsmith/organization-workspace-operations).
* You have set the [gateway environment variables](/langsmith/llm-gateway-quickstart#1-set-environment-variables) in your terminal.

## Supported clients

* Claude Code CLI
* Codex CLI
* Gemini CLI
* Deep Agents

## Claude Code CLI

No extra configuration beyond the [environment variables](/langsmith/llm-gateway-quickstart#1-set-environment-variables). Run:

```bash theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
claude
```

Claude Code will use `ANTHROPIC_BASE_URL` and `ANTHROPIC_API_KEY` from your environment automatically.

<Warning>
  Claude Desktop desktop plugins break when the gateway is configured. Claude users on a paid plan (Plus, Max) are not yet supported.
</Warning>

## Codex CLI

Codex requires a TOML configuration file in addition to environment variables.

Add this to `~/.codex/config.toml`:

```toml theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
model_provider = "langsmith-gateway"

[model_providers.langsmith-gateway]
name = "LangSmith Gateway"
base_url = "https://gateway.smith.langchain.com/openai/v1"
env_key = "LANGSMITH_API_KEY"
supports_websockets = false
```

Make sure the `LANGSMITH_API_KEY` environment variable is [set](/langsmith/llm-gateway-quickstart#1-set-environment-variables), then run:

```bash theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
codex
```

<Warning>
  Codex Desktop plugins break when the gateway is configured. The TOML configuration forces authentication through the gateway, so OpenAI no longer handles plugin auth directly.
</Warning>

## Gemini CLI

No extra configuration beyond the [environment variables](/langsmith/llm-gateway-quickstart#1-set-environment-variables). Run:

```bash theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
gemini
```

Gemini CLI will use `GOOGLE_GEMINI_BASE_URL` and `GEMINI_API_KEY` from your environment automatically.

## Deep Agents

No extra configuration beyond the [environment variables](/langsmith/llm-gateway-quickstart#1-set-environment-variables). For details, refer to the [provider selection docs](/oss/python/deepagents/code/providers#provider-reference). Run:

```bash theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
deepagents
```

## Python SDK usage

If you call LLM providers from Python scripts rather than through a coding agent, swap the `base_url`:

<Tabs>
  <Tab title="OpenAI SDK">
    ```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
    import os

    from openai import OpenAI

    client = OpenAI(
        base_url="https://gateway.smith.langchain.com/openai/v1",
        api_key=os.environ["LANGSMITH_API_KEY"],
    )
    response = client.chat.completions.create(
        model="gpt-4o-mini",
        messages=[{"role": "user", "content": "Say hello!"}],
    )
    print(response.choices[0].message.content)
    ```
  </Tab>

  <Tab title="Anthropic SDK">
    ```python theme={"theme":{"light":"catppuccin-latte","dark":"catppuccin-mocha"}}
    import os

    import anthropic

    client = anthropic.Anthropic(
        base_url="https://gateway.smith.langchain.com/anthropic",
        api_key=os.environ["LANGSMITH_API_KEY"],
    )
    message = client.messages.create(
        model="claude-sonnet-4-6",
        max_tokens=1024,
        messages=[{"role": "user", "content": "Say hello!"}],
    )
    print(message.content[0].text)
    ```
  </Tab>
</Tabs>

## Company-wide deployment

For organizations rolling the gateway out to all developers, distribute the configuration through MDM (mobile device management) or a shared shell profile. The key pieces to distribute are:

1. The gateway base URL (`https://gateway.smith.langchain.com`).
2. A workspace-scoped [LangSmith API key](/langsmith/create-account-api-key) per user (or per team, depending on your policy granularity).
3. The Codex `config.toml` if your organization uses Codex.

Provider API keys stay centralized in LangSmith workspace secrets.

## Verify the setup

After configuring a coding agent, make a test call and confirm that:

1. The call succeeds (the agent gets a response).
2. A trace appears in the `gateway` and `gateway-<short_api_key>-<api_key_id>` tracing projects in your LangSmith workspace.

If the call fails with a `403`, check that your API key's role includes `gateway:invoke` and `workspaces:read`. If it fails with a `400` mentioning a missing provider key, ask your org admin to add the provider's key to workspace secrets.

## Next steps

* [Spend policies](/langsmith/llm-gateway-spend-policies): set cost limits on developer LLM usage.
* [Traces, Engine, and access control](/langsmith/llm-gateway-access): understand where gateway traces appear.

***

<div className="source-links">
  <Callout icon="terminal-2">
    [Connect these docs](/use-these-docs) to Claude, VSCode, and more via MCP for real-time answers.
  </Callout>

  <Callout icon="edit">
    [Edit this page on GitHub](https://github.com/langchain-ai/docs/edit/main/src/langsmith/llm-gateway-coding-agents.mdx) or [file an issue](https://github.com/langchain-ai/docs/issues/new/choose).
  </Callout>
</div>