Claude Code CLI Setup

Create a Gate.AI API Key

1. Open gate.ai → Dashboard → API Keys, create and copy a key starting with sk-v1-…
2. Confirm your account has sufficient balance

Network connectivity check

Replace GATEAI_API_KEY with your key:

1export GATEAI_API_KEY="sk-v1-xxxxxxxxxxxxxxxx"

Anthropic-compatible endpoint (Claude Code):

1curl -s -o /dev/null -w "%{http_code}" \2  -H "x-api-key: $GATEAI_API_KEY" \3  -H "content-type: application/json" \4  -H "anthropic-version: 2023-06-01" \5  -d '{"model":"anthropic/claude-sonnet-4.6","max_tokens":16,"messages":[{"role":"user","content":"hi"}]}' \6  https://api.gate.ai/anthropic/v1/messages

• Returns 200: gateway is reachable; proceed with install and configuration
• Returns 401: invalid or expired key — check the dashboard
• Connection timeout: check local network or DNS; do not use https://api.gate.ai/v1

Install Claude Code CLI

macOS / Linux / WSL (recommended):

1curl -fsSL https://claude.ai/install.sh | bash

Or with npm:

1npm install -g @anthropic-ai/claude-code

If npm install fails

This is usually caused by registry.npmjs.org timeouts or instability — unrelated to the Gate.AI gateway. Try one of the following:

Option A: one-off install with a mirror (try first)

1npm install -g @openai/codex --registry=https://registry.npmmirror.com2npm install -g @anthropic-ai/claude-code --registry=https://registry.npmmirror.com

Option B: set mirror globally

1# Use npmmirror (common in China)2npm config set registry https://registry.npmmirror.com34# Verify registry5npm config get registry67# Reinstall8npm install -g @openai/codex9npm install -g @anthropic-ai/claude-code

Option C: current shell session only

1export NPM_CONFIG_REGISTRY=https://registry.npmmirror.com2npm install -g @openai/codex

Configure model

Replace all sk-v1-your-key placeholders with your real key. No need to repeat unless you change the key or model.

1mkdir -p ~/.claude23cat > ~/.claude/settings.json <<'EOF'4{5  "env": {6    "ANTHROPIC_BASE_URL": "https://api.gate.ai/anthropic",7    "ANTHROPIC_API_KEY": "sk-v1-your-key",8    "ANTHROPIC_MODEL": "anthropic/claude-sonnet-4.6"9  },10  "includeCoAuthoredBy": false11}12EOF1314# Clear old Anthropic / proxy sessions to avoid auth conflicts15claude /logout 2>/dev/null || true16unset ANTHROPIC_AUTH_TOKEN

Store credentials only in settings.json. If ~/.zshrc still has ANTHROPIC_AUTH_TOKEN or a duplicate ANTHROPIC_API_KEY, comment or remove them.

Verify setup

In your project directory, start the CLI for AI-assisted coding in the terminal.

Run claude

1claude

First check: run /status and confirm Base URL is https://api.gate.ai/anthropic and auth token is ANTHROPIC_API_KEY.

Advanced configuration

Auth conflict

If you see Auth conflict: Both a token (ANTHROPIC_AUTH_TOKEN) and an API key (ANTHROPIC_API_KEY) are set:

Alternative configuration

Shell environment variables (do not duplicate settings.json):

1export ANTHROPIC_BASE_URL="https://api.gate.ai/anthropic"2export ANTHROPIC_API_KEY="sk-v1-xxxxxxxxxxxxxxxx"3export ANTHROPIC_MODEL="anthropic/claude-sonnet-4.6"

Project-level settings (share structure; never commit real keys):

Model environment variables

Add to the env block in settings.json:

Full list: model catalog. If auto fails, use an explicit model ID.

Restore direct Anthropic (optional)

1env -u ANTHROPIC_BASE_URL -u ANTHROPIC_API_KEY claude

Troubleshooting

Common Base URL mistakes

1# ❌ Wrong (missing /anthropic prefix, 404)2https://api.gate.ai/v13https://api.gate.ai/v1/chat/completions45# ❌ Wrong (do not use full API path in Claude config)6https://api.gate.ai/anthropic/v1/messages7https://api.gate.ai/anthropic/messages89# ✅ Correct10https://api.gate.ai/anthropic/v111https://api.gate.ai/anthropic          # Claude Code CLI12https://api.gate.ai/anthropic/v1/messages  # curl key check (Anthropic)13https://api.gate.ai/openai/v1/responses    # curl key check (Codex)

• /openai/v1/chat/completions works, but Codex must use the Responses API. On 404 for /responses, check wire_api = "responses", not the URL.
• /openai/v1/models does not validate keys (invalid keys may still return 200). Use the curl endpoints above to verify keys.

Quick reference