# Claude Code CLI
> Connect the Claude Code CLI to the LMU AI API: settings.json config, model selection, Chinese LLMs, enabling the 1M context, and common error fixes.
URL: https://docs.lmuai.ai/docs/tools/claude-code
Claude Code is Anthropic's official AI coding agent — it runs in your terminal and lets you write, debug, and refactor code in natural language.
If you would rather not edit `settings.json` by hand, use [**CC Switch one-click import**](/docs/tools/cc-switch): click the "Import to CCS" button in the key list in the LMU AI console, and the Base URL and key are configured automatically.
***
## Step 1 — Install Node.js [#step-1--install-nodejs]
> Skip this if you already have it; check with `node -v`.
You need **Node.js 18+** — download: [https://nodejs.org/en/download](https://nodejs.org/en/download)
```bash
# Verify after installing
node -v
# A version number (e.g. v24.4.1) means it installed correctly
```
```powershell
# During install, check "Automatically install the necessary tools"
# Reopen PowerShell after installing and verify
node -v
```
If you installed it before but the version is too old or the PATH is broken, reinstall and reopen the terminal.
***
## Step 2 — Install Claude Code [#step-2--install-claude-code]
```bash
npm install -g @anthropic-ai/claude-code
```
```powershell
npm install -g @anthropic-ai/claude-code
```
Verify the install:
```bash
claude --version
```
```bash
npm config set registry https://registry.npmmirror.com
npm install -g @anthropic-ai/claude-code
```
***
## Step 3 — Configure the LMU AI API [#step-3--configure-the-lmu-ai-api]
Claude Code configures a custom API endpoint through `settings.json`.
### Config file location [#config-file-location]
```bash
~/.claude/settings.json
```
```powershell
C:\Users\YourUsername\.claude\settings.json
```
### Create the config file [#create-the-config-file]
```bash
mkdir -p ~/.claude && touch ~/.claude/settings.json
```
```powershell
mkdir "$env:USERPROFILE\.claude" -Force
New-Item "$env:USERPROFILE\.claude\settings.json" -Force
```
### Write the config [#write-the-config]
Open `settings.json` in a text editor and add the following (replace the key with yours):
```json
{
"env": {
"ANTHROPIC_BASE_URL": "https://api.lmuai.ai",
"ANTHROPIC_AUTH_TOKEN": "sk-your-lmu-ai-api-key",
"API_TIMEOUT_MS": "3000000",
"CLAUDE_CODE_ATTRIBUTION_HEADER": "0"
}
}
```
* Set `ANTHROPIC_AUTH_TOKEN` to the API key you generated in the LMU AI console (starts with `sk-`)
* Do not use an official Anthropic API key
* If you configured an official key before, clear the old config first, then write this in
* Setting `CLAUDE_CODE_ATTRIBUTION_HEADER` to `"0"` turns off the source-attribution header on requests, which helps cache hits and token efficiency
***
## Enabling the 1M context in Claude Code (optional) [#enabling-the-1m-context-in-claude-code-optional]
Claude Opus 4.8 / Sonnet 5 support a **1M-token long context window** (200K by default), useful for huge repos, long logs, and cross-file refactors. Add the `[1M]` suffix to the model ID to enable it.
### Option A — manual /model switch (recommended for occasional use) [#option-a--manual-model-switch-recommended-for-occasional-use]
**Keep a minimal settings.json** (just `ANTHROPIC_BASE_URL` + `ANTHROPIC_AUTH_TOKEN`), start Claude Code, and type in the prompt:
```
/model claude-opus-5[1M]
```
or:
```
/model claude-sonnet-5[1M]
```
This switches to the 1M version with no config changes. It applies to the current session and reverts on exit.
### Option B — a standing default in settings.json (recommended for daily long-context use) [#option-b--a-standing-default-in-settingsjson-recommended-for-daily-long-context-use]
Add two lines to the `env` section of `settings.json` so `/model opus` / `/model sonnet` default to the 1M version:
```json
{
"env": {
"ANTHROPIC_BASE_URL": "https://api.lmuai.ai",
"ANTHROPIC_AUTH_TOKEN": "sk-your-lmu-ai-api-key",
"ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-5[1M]",
"ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-sonnet-5[1M]",
"API_TIMEOUT_MS": "3000000",
"CLAUDE_CODE_ATTRIBUTION_HEADER": "0"
}
}
```
Save and restart Claude Code; selecting opus / sonnet in the `/model` command now uses the 1M version automatically.
**Field reference**
| Field | Purpose |
| -------------------------------- | ------------------------------------------------------------------------------------ |
| `ANTHROPIC_DEFAULT_OPUS_MODEL` | The model ID actually sent when you pick `opus` in Claude Code |
| `ANTHROPIC_DEFAULT_SONNET_MODEL` | Same, for `sonnet` |
| `[1M]` suffix | Enables the model's 1M-token long-context mode; without it, the default 200K applies |
* **Different billing**: the 1M context mode is tiered by Anthropic pricing, and the per-token price is usually **higher than the default 200K mode**, so long-text tasks cost significantly more — do not leave it on unless needed
* **Only some models support it**: the Opus / Sonnet mainline models `claude-opus-5`, `claude-fable-5`, `claude-opus-4-8`, `claude-opus-4-7`, `claude-sonnet-5` support the `[1M]` suffix; the Haiku series and older models do not. The [Model Gallery](/docs/guide/models) list is the source of truth
* **Choose Option A for occasional use, Option B for regular use** — pick one
***
## Using Chinese models (optional) [#using-chinese-models-optional]
LMU AI supports Chinese LLMs (e.g. the Qwen series). By setting `model` in `settings.json`, **you skip the manual `/model` switch on every launch** and use the specified model directly.
```json
{
"env": {
"ANTHROPIC_BASE_URL": "https://api.lmuai.ai",
"ANTHROPIC_AUTH_TOKEN": "sk-your-lmu-ai-api-key",
"CLAUDE_CODE_ATTRIBUTION_HEADER": "0"
},
"model": "qwen3.8-max-preview",
"effortLevel": "medium"
}
```
**Field reference:**
| Field | Description |
| -------------------------------- | ------------------------------------------------------------------------------------------------------------------- |
| `model` | The default model, loaded on startup — no manual switch each time |
| `effortLevel` | Reasoning effort: `low` / `medium` / `high`; `medium` is recommended for Chinese models |
| `CLAUDE_CODE_ATTRIBUTION_HEADER` | Set to `"0"` to turn off the source-attribution header, which helps cache hits, token efficiency, and compatibility |
* `qwen3.8-max-preview` — Qwen 3.8 Max Preview (latest)
* `qwen3.7-max` — Qwen 3.7 Max
* `glm-5.2` — Zhipu GLM-5.2
* `deepseek-v4-pro` — DeepSeek V4 Pro
* `kimi-k3` — Kimi K3
The **Available Models** list in the LMU AI console is the source of truth.
| | Chinese models | Official Claude models |
| --------------------- | -------------- | ---------------------- |
| Cost | Lower | Higher |
| Chinese comprehension | Excellent | Good |
| Coding ability | Excellent | Excellent |
| Default-model setting | ✅ Supported | ✅ Supported |
***
## Step 4 — Launch Claude Code [#step-4--launch-claude-code]
Open a terminal in your project directory and run:
```bash
claude
```
### No-approval mode (recommended) [#no-approval-mode-recommended]
```bash
claude --dangerously-skip-permissions
```
> In this mode Claude Code runs commands automatically without confirming each step — handy inside a project directory.
***
## Verify the config [#verify-the-config]
After launching, type in the Claude Code interface:
```
/status
```
It shows the current:
* Model name
* API Base URL (should show `https://api.lmuai.ai`)
* API key status
If the Base URL is correct, the config is working.
***
## Common issues [#common-issues]
### 401 Unauthorized [#401-unauthorized]
**Cause:** the API key is wrong, or the request still went to Anthropic's official endpoint.
**Fix:**
1. Confirm `ANTHROPIC_AUTH_TOKEN` in `settings.json` is your LMU AI key
2. Check whether a shell environment variable is overriding the config
```bash
echo $ANTHROPIC_BASE_URL
```
```powershell
echo $env:ANTHROPIC_BASE_URL
```
3. Reopen the terminal, then launch Claude Code again
### stream disconnected [#stream-disconnected]
**Cause:** an unstable local network, or a VPN / proxy / system proxy is on — a proxy rotating the IP breaks the connection.
**Fix:** turn off the VPN / proxy / system proxy and retry. The LMU AI overseas gateway connects directly; a direct connection is fastest and most stable.
### 503 No available accounts [#503-no-available-accounts]
**Cause:** usually `~/.zshrc` or `~/.bashrc` sets global env vars like `ANTHROPIC_AUTH_TOKEN` / `ANTHROPIC_BASE_URL`, which override `settings.json`.
**Fix:** remove those lines from your shell config, or move them to a separate file you only `source` when launching Claude Code. See [FAQ · Issue 6](/docs/guide/faq#issue-6).
### Response timeouts [#response-timeouts]
**Cause:** the default timeout is short.
**Fix:** confirm `API_TIMEOUT_MS` is set to `3000000` (50 minutes) so long tasks are not cut off.
***
## Tips [#tips]
* Launch Claude Code from the project root so it picks up the project structure automatically
* Pass an instruction directly, e.g. `claude "refactor this function for me"`
* `Ctrl+C` interrupts the current task, `/exit` quits
* Type `?` or `/help` to see all commands
***
## Using /goal to make Claude Code work until the goal is met (optional) [#using-goal-to-make-claude-code-work-until-the-goal-is-met-optional]
`/goal` is a built-in Claude Code slash command (available since **v2.1.139**, May 2026) that sets a **completion condition** for the current session. Once set, Claude **keeps going across turns on its own** until the condition is judged met, rather than stopping when "it feels done." It is especially handy on LMU AI for long, run-to-completion tasks (large-repo refactors, implementing to an acceptance spec, clearing an issue backlog).
> Older versions do not have this command. If it reports an unknown command, upgrade Claude Code to the latest version first.
### Basic usage [#basic-usage]
| Command | Purpose |
| ------------------------------ | --------------------------------------------------------------------------------------------------- |
| `/goal ` | Set a goal; Claude starts a turn immediately and continues automatically until the condition is met |
| `/goal` | Show the status and progress of the current (or most recent) goal |
| `/goal clear` | Clear the current goal early (`stop` / `off` / `reset` / `cancel` / `none` are equivalent) |
A session has **only one goal at a time**; setting a new one replaces the old and starts a new turn immediately.
### How it judges "met" [#how-it-judges-met]
At the end of each turn, Claude Code hands **your completion condition + this turn's conversation** to a small, fast model (**Haiku** by default) to judge. The judge **only looks at evidence already in the conversation** — test output, build logs, file diffs — and **will not re-run your whole CI behind your back**. So write the condition in a form where Claude can "produce evidence" in the conversation.
The goal is **cleared automatically** in any of these cases:
* the condition is judged **met**;
* the model judges the condition **impossible to satisfy**;
* a turn **hits an error that needs your intervention**.
### Writing a good goal condition [#writing-a-good-goal-condition]
* **Use a verifiable end state**: e.g. "`npm test` exits 0", "`tsc --noEmit` reports no errors" — not subjective descriptions like "make the code prettier";
* **Let evidence land in the conversation**: print test / build results each turn so the judge model can see them;
* **Scope it + cap the turns**: e.g. "only change files under `src/auth/`, stop after at most 20 turns" to avoid spinning;
* Requires a **trusted workspace (hooks enabled)** to take effect.
Each extra `/goal` turn spends another round of tokens (plus a little Haiku overhead for the per-turn judging). When running a long goal on LMU AI, write a clear **end state** and a **max turn count** into the condition so it does not burn quota spinning on an unverifiable goal. For a "self-correct repeatedly until checks pass" loop, pair `/goal` with `/loop`.