Setup Claude Code with Enterprise LLMs

Overview

Claude Code is Anthropic’s official CLI for interacting with Claude models. In enterprise environments, getting it to work requires navigating:

• Corporate HTTPS proxies that intercept and re-sign TLS connections
• Custom CA certificates that need to be trusted by Node.js
• Proxy authentication requirements
• Network access restrictions to external AI APIs

This guide covers the configuration steps to get Claude Code running in a managed enterprise environment.

Prerequisites

• Node.js 18+ installed
• Claude Code installed: npm install -g @anthropic-ai/claude-code
• Your corporate CA certificate (.pem or .crt format)
• Proxy server address (if required)
• An Anthropic API key (or access to an Anthropic-compatible endpoint)

Step 1: Identify Your Network Environment

First, determine what you’re working with:

# Check for corporate proxy
env | grep -i proxy
curl -v https://api.anthropic.com 2>&1 | head -30

# Inspect the certificate being presented
echo | openssl s_client -connect api.anthropic.com:443 2>/dev/null | \
  openssl x509 -noout -issuer -subject

If the issuer is your organisation (not Anthropic or a public CA), SSL inspection is active.

Step 2: Export the Corporate CA Certificate

From a browser (Chrome/Edge):

1. Navigate to any HTTPS internal site
2. Click the padlock → Certificate → Details → Copy to File
3. Export as Base-64 encoded X.509 (.CER)

Via command line:

# Extract from a known internal HTTPS site
echo | openssl s_client -connect internal.example.com:443 -showcerts 2>/dev/null | \
  awk '/-----BEGIN CERTIFICATE-----/,/-----END CERTIFICATE-----/' > corporate-ca.pem

# On Windows (PowerShell) - export from cert store
$cert = Get-ChildItem Cert:\LocalMachine\Root | Where-Object {$_.Subject -like "*YourCorp*"}
Export-Certificate -Cert $cert -FilePath corporate-ca.pem -Type CERT

Step 3: Configure Node.js to Trust the Corporate CA

Claude Code runs on Node.js, which uses its own bundled CA list by default. Override this with:

# Set permanently in your shell profile (.bashrc / .zshrc)
export NODE_EXTRA_CA_CERTS=/path/to/corporate-ca.pem

# Verify it works
node -e "const https = require('https'); https.get('https://api.anthropic.com', r => console.log(r.statusCode))"

Step 4: Configure Proxy Settings

# In your shell profile
export HTTPS_PROXY=http://proxy.internal.example.com:8080
export HTTP_PROXY=http://proxy.internal.example.com:8080
export NO_PROXY=localhost,127.0.0.1,10.0.0.0/8,.internal.example.com

# If proxy requires authentication
export HTTPS_PROXY=http://username:password@proxy.internal.example.com:8080

Security note: Avoid embedding credentials in NO_PROXY if possible. Use a credentials manager or discuss with your proxy team about IP-based authentication for developer workstations.

Step 5: Configure Claude Code

# Set your API key
export ANTHROPIC_API_KEY=sk-ant-...

# Or configure via claude CLI
claude config set api-key sk-ant-...

# Test the connection
claude --version
echo "Hello" | claude

Step 6: Verify the Setup

# Quick connectivity test
claude -p "What is 2+2?"

# If you get SSL errors, double-check:
echo $NODE_EXTRA_CA_CERTS
echo $HTTPS_PROXY
openssl verify -CAfile $NODE_EXTRA_CA_CERTS /dev/stdin <<< "$(echo | openssl s_client -connect api.anthropic.com:443 2>/dev/null | openssl x509)"

Using an Internal/Alternative Endpoint

Some organisations proxy or cache LLM API calls through an internal gateway. Claude Code supports custom base URLs:

# Point to an internal Anthropic-compatible endpoint
export ANTHROPIC_BASE_URL=https://llm-gateway.internal.example.com

# Or configure via claude
claude config set api-url https://llm-gateway.internal.example.com

The endpoint must implement the Anthropic Messages API format.

Persistent Configuration

Add these to your shell profile for persistence:

# ~/.bashrc or ~/.zshrc

# Claude Code enterprise configuration
export ANTHROPIC_API_KEY="sk-ant-..."
export NODE_EXTRA_CA_CERTS="/home/username/.certs/corporate-ca.pem"
export HTTPS_PROXY="http://proxy.internal.example.com:8080"
export NO_PROXY="localhost,127.0.0.1,.internal.example.com"

Troubleshooting

Security Considerations

When running AI coding tools in enterprise environments:

1. Understand what gets sent — Claude Code sends your code and queries to the API. Ensure this complies with your data classification policies
2. API key management — Store API keys in a secrets manager, not in plaintext config files
3. Audit logging — Some organisations require logging of AI API calls; check with your security team
4. Data residency — Verify Anthropic’s data processing regions align with your compliance requirements (GDPR, Privacy Act, etc.)