YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context
Rust agentsagiaiai-memoryaudio-recordingcomputer-visionhermeshermes-agentllmlocal-ailocal-firstmachine-learningmcpmultimodalopenclawprivacyrewindscreen-recordingspeech-to-textycombinator
/**
* Tests for Vertex AI proxy
*
* To run these tests locally:
* 1. Set up your service account JSON in .dev.vars:
* VERTEX_SERVICE_ACCOUNT_JSON='{"type":"service_account",...}'
* VERTEX_PROJECT_ID='your-project-id'
* VERTEX_REGION='us-east5'
*
* 2. Start the worker: npm run dev
*
* 3. Run tests: npx bun test src/test/vertex-proxy.test.ts
* Or manually with curl (see examples below)
*/
import { describe, it, expect } from 'bun:test';
const HOST = process.env.TEST_HOST || 'http://localhost:8787';
async function fetchRunningWorker(path: string): Promise<Response | null> {
try {
return await fetch(`${HOST}${path}`);
} catch (error) {
if (String(error).includes('ConnectionRefused') || String(error).includes('Unable to connect')) {
console.log('Skipping test - worker is not running');
return null;
}
throw error;
}
}
describe('Vertex AI Proxy', () => {
// Skip tests if no service account is configured
const hasCredentials = process.env.VERTEX_SERVICE_ACCOUNT_JSON || process.env.SKIP_CREDENTIAL_CHECK;
it('should return health check', async () => {
const response = await fetchRunningWorker('/test');
if (!response) return;
expect(response.status).toBe(200);
const text = await response.text();
expect(text).toContain('ai proxy is working');
});
it('should handle /v1/messages endpoint', async () => {
if (!hasCredentials) {
console.log('Skipping test - no credentials configured');
return;
}
const response = await fetch(`${HOST}/v1/messages`, {
method: 'POST',
headers: {
'Content-Type': 'application/json',
Authorization: 'Bearer test-token',
},
body: JSON.stringify({
model: 'claude-3-5-sonnet-v2@20241022',
max_tokens: 100,
messages: [
{
role: 'user',
content: 'Say "hello" and nothing else.',
},
],
}),
});
console.log('Response status:', response.status);
const data = await response.json();
console.log('Response data:', JSON.stringify(data, null, 2));
// If credentials are invalid, we'll get a 500 with error message
// If valid, we should get 200 with response
expect([200, 500]).toContain(response.status);
});
it('should handle streaming requests', async () => {
if (!hasCredentials) {
console.log('Skipping test - no credentials configured');
return;
}
const response = await fetch(`${HOST}/v1/messages`, {
method: 'POST',
headers: {
'Content-Type': 'application/json',
Authorization: 'Bearer test-token',
},
body: JSON.stringify({
model: 'claude-3-5-sonnet-v2@20241022',
max_tokens: 100,
stream: true,
messages: [
{
role: 'user',
content: 'Count from 1 to 3.',
},
],
}),
});
console.log('Streaming response status:', response.status);
expect([200, 500]).toContain(response.status);
if (response.status === 200) {
const reader = response.body?.getReader();
if (reader) {
const chunks: string[] = [];
const decoder = new TextDecoder();
while (true) {
const { done, value } = await reader.read();
if (done) break;
chunks.push(decoder.decode(value, { stream: true }));
}
console.log('Received chunks:', chunks.length);
expect(chunks.length).toBeGreaterThan(0);
}
}
});
});
/*
Manual test commands:
# Start the worker
cd packages/ai-gateway
npm run dev
# Test health check
curl http://localhost:8787/test
# Test /v1/messages (non-streaming)
curl -X POST http://localhost:8787/v1/messages \
-H "Content-Type: application/json" \
-H "Authorization: Bearer test-token" \
-d '{
"model": "claude-3-5-sonnet-v2@20241022",
"max_tokens": 100,
"messages": [
{"role": "user", "content": "Say hello"}
]
}'
# Test /v1/messages (streaming)
curl -X POST http://localhost:8787/v1/messages \
-H "Content-Type: application/json" \
-H "Authorization: Bearer test-token" \
-d '{
"model": "claude-3-5-sonnet-v2@20241022",
"max_tokens": 100,
"stream": true,
"messages": [
{"role": "user", "content": "Count from 1 to 5"}
]
}'
# Test with Agent SDK (in your app)
export CLAUDE_CODE_USE_VERTEX=1
export ANTHROPIC_VERTEX_BASE_URL=http://localhost:8787
export CLAUDE_CODE_SKIP_VERTEX_AUTH=1
# Then run your Agent SDK code
*/