Merge pull request #29 from navapbc/fg/vertexai-provider

feat: add Vertex AI Anthropic integration and setup documentation

Author: fg-nava

mastra-test-app/.env.example

@@ -1,3 +1,18 @@
+# LLM API Key Configuration
 OPENAI_API_KEY=sk-proj-123
 ANTHROPIC_API_KEY=sk-ant-123
-EXA_API_KEY=abc-123
+EXA_API_KEY=abc-123
+GOOGLE_GENERATIVE_AI_API_KEY=AI...
+
+# Google Vertex AI Configuration
+GOOGLE_VERTEX_LOCATION=us-east5
+GOOGLE_VERTEX_PROJECT=project-id
+GOOGLE_APPLICATION_CREDENTIALS=path/to/vertex-ai-credentials.json
+
+# Database Configuration
+DATABASE_URL="postgresql://..."
+
+# Authentication Configuration
+MASTRA_JWT_SECRET=jwt-secret
+MASTRA_APP_PASSWORD=your-password
+MASTRA_JWT_TOKEN=jwt-token

mastra-test-app/.gitignore

@@ -27,3 +27,6 @@ mastra.db
 mastra.db-*
 mastra-memory.db
 mastra-memory.db-*
+
+# Google Vertex AI
+vertex-ai-credentials.json

mastra-test-app/docs/VERTEX_AI_ANTHROPIC_SETUP.md

@@ -0,0 +1,147 @@
+# Vertex AI Anthropic Setup Guide
+
+This guide shows how to set up Google Vertex AI with Anthropic's Claude models for your Node.js application.
+
+## Prerequisites
+
+- Google Cloud Platform account with billing enabled
+- A GCP project
+- Node.js application with the AI SDK
+
+## Step 1: Install Dependencies
+
+Add the required packages to your project:
+
+```bash
+pnpm add @ai-sdk/google-vertex google-auth-library
+```
+
+Your `package.json` should include these versions (or newer):
+
+```json
+{
+  "dependencies": {
+    "@ai-sdk/google-vertex": "^2.2.27",
+    "google-auth-library": "^10.2.1"
+  }
+}
+```
+
+## Step 2: Configure Google Cloud
+
+### Install and configure gcloud CLI:
+
+```bash
+# Install gcloud CLI (if not already installed)
+# Follow instructions at: https://cloud.google.com/sdk/docs/install
+
+# Set your project
+gcloud config set project YOUR-PROJECT-ID
+
+# Enable Vertex AI API
+gcloud services enable aiplatform.googleapis.com
+
+# Authenticate
+gcloud auth login
+gcloud auth application-default login
+```
+
+## Step 3: Enable Anthropic Models
+
+1. Go to the [Vertex AI Model Garden](https://console.cloud.google.com/vertex-ai/model-garden)
+2. Navigate to Anthropic models or use this direct link for Claude Sonnet 4: https://console.cloud.google.com/vertex-ai/publishers/anthropic/model-garden/claude-sonnet-4
+3. Click "Enable" for the models you want to use
+4. Complete the access request form
+5. Wait 1-2 business days for approval
+
+## Step 4: Set Environment Variables
+
+Add these variables to your `.env` file:
+
+```bash
+# Google Vertex AI Configuration
+GOOGLE_VERTEX_LOCATION=us-east5
+GOOGLE_VERTEX_PROJECT=your-project-id
+GOOGLE_APPLICATION_CREDENTIALS=${PWD}/vertex-ai-credentials.json
+```
+
+Note: Use `us-east5` as the location since it has the best model availability.
+
+## Step 5: Create Service Account Credentials
+
+### Option A: Using Google Cloud Console
+
+1. Go to [Google Cloud Console IAM](https://console.cloud.google.com/iam-admin/serviceaccounts)
+2. Create a new service account or use an existing one
+3. Add the `Vertex AI User` role
+4. Generate a JSON key file
+5. Save it as `vertex-ai-credentials.json` in your project root
+
+### Option B: Using gcloud CLI
+
+Create a service account and generate credentials using the gcloud CLI:
+
+```bash
+# Create a service account
+gcloud iam service-accounts create vertex-ai \
+  --display-name="Vertex AI Service Account"
+
+# Add the Vertex AI User role
+gcloud projects add-iam-policy-binding YOUR-PROJECT-ID \
+  --member="serviceAccount:[email protected]" \
+  --role="roles/aiplatform.user"
+
+# Generate and download the credentials JSON file
+gcloud iam service-accounts keys create vertex-ai-credentials.json \
+  [email protected]
+```
+
+Replace `YOUR-PROJECT-ID` with your actual Google Cloud project ID.
+
+## Step 6: Use in Your Code
+
+Import and use the Vertex AI Anthropic provider:
+
+```typescript
+import { vertexAnthropic } from '@ai-sdk/google-vertex/anthropic';
+
+// Use in your agent or application
+const model = vertexAnthropic('claude-sonnet-4@20250514');
+```
+
+## Step 7: Test the Setup
+
+Test your configuration with a curl request:
+
+```bash
+curl -X POST \
+  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
+  -H "Content-Type: application/json; charset=utf-8" \
+  -d '{
+    "anthropic_version": "vertex-2023-10-16",
+    "messages": [
+      {
+        "role": "user",
+        "content": "Hello"
+      }
+    ],
+    "max_tokens": 100
+  }' \
+  "https://aiplatform.googleapis.com/v1/projects/YOUR-PROJECT-ID/locations/global/publishers/anthropic/models/claude-sonnet-4@20250514:streamRawPredict"
+```
+
+Replace `YOUR-PROJECT-ID` with your actual Google Cloud project ID.
+
+## Troubleshooting
+
+**Model not found errors**: Verify you have access to the model in the specified region and that it's enabled in Model Garden.
+
+**Authentication errors**: Check that your service account has the correct permissions and that the credentials file path is correct.
+
+**Quota errors**: Request quota increases in the [Cloud Console](https://console.cloud.google.com/iam-admin/quotas).
+
+## Helpful Resources
+
+- [Vertex AI Documentation](https://cloud.google.com/vertex-ai/docs)
+- [Anthropic Claude Code Vertex AI Guide](https://docs.anthropic.com/en/docs/claude-code/google-vertex-ai)
+- [AI SDK Google Vertex Documentation](https://sdk.vercel.ai/providers/ai-sdk-providers/google-vertex)

mastra-test-app/package.json

@@ -28,6 +28,7 @@
     "@ai-sdk/anthropic": "^1.2.12",
     "@ai-sdk/google": "^1.2.22",
     "@ai-sdk/openai": "^1.3.23",
+    "@ai-sdk/google-vertex": "^2.2.27",
     "@inquirer/prompts": "^7.7.1",
     "@mastra/core": "^0.13.2",
     "@mastra/libsql": "^0.13.2",
@@ -40,6 +41,7 @@
     "csv-parse": "^6.1.0",
     "dotenv": "^17.2.1",
     "exa-mcp-server": "^2.0.3",
+    "google-auth-library": "^10.2.1",
     "jsonwebtoken": "^9.0.2",
     "pg": "^8.16.3",
     "pino-pretty": "^13.0.0",