API quickstart

From zero to your first Ostra Decision and Ostra Pro calls in five minutes — key, credit, cURL, Python, TypeScript and MCP.

1. Get a key and some credit

Sign in to the TestRelic Console with your TestRelic account (the same login as the TestRelic platform). A new account gets its own workspace.

Add your billing details and at least $5 of credit. The Console is pay as you go and prepaid: every request made with an API key needs credit on the balance, and the Console has no subscription. See Billing and credits.

Open API keys → Create key. Leave access on Full access, copy the key (ostk_live_…) and store it in your secrets manager. It is shown only once.

Every sample on this site reads the key from this variable
export OSTRA_API_KEY="ostk_live_..."

Keep keys out of code

Never commit a key or put it in client-side code. Use environment variables or your CI's secret store, and create one key per system so you can revoke one without breaking the rest.

2. Check the key

GET /models is free and works with any key:

curl https://ostra.testrelic.ai/api/v1/models -H "Authorization: Bearer $OSTRA_API_KEY"

A 200 with two models means you're set. A 401 means the key is wrong or revoked; see Authentication.

3. Triage a failing test (Ostra Decision)

curl https://ostra.testrelic.ai/api/v1/predict \
  -H "Authorization: Bearer $OSTRA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "ostra-lite",
    "task": "triage",
    "input": {
      "test_id": "checkout.spec.ts > applies coupon",
      "context": {
        "framework": "playwright",
        "action": "click",
        "error_message": "TimeoutError: locator.click: Timeout 30000ms exceeded.",
        "memory": { "runs_50": 40, "failed_count_50": 6, "flake_rate_50": 0.15 }
      }
    }
  }'

output.decision is one of flaky, likely_flaky, product_bug, test_bug, environment, selector_drift or defer, with its probability in p. What to do with each answer: Failure triage.

4. Ask Ostra Pro

Ostra Pro is OpenAI-compatible: point any OpenAI client at https://ostra.testrelic.ai/api/v1.

curl https://ostra.testrelic.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OSTRA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "ostra-pro",
    "reasoning_effort": "low",
    "messages": [
      { "role": "user", "content": "Write a Playwright test that logs in and checks the dashboard title." }
    ]
  }'

The first call after a quiet period can wait

Ostra Pro scales to zero after 5 idle minutes. The next request starts it, which usually takes under a minute. Streaming requests simply wait; non-streaming ones may get 503 with Retry-After. Call POST /warm first if you're about to send a burst. Details: Cold starts.

5. Give your coding agent the tools

Connect the Ostra MCP server and your agent can call every model directly. For Claude Code:

claude mcp add --transport http ostra https://ostra.testrelic.ai/mcp --header "Authorization: Bearer $OSTRA_API_KEY"

Cursor, VS Code and other clients: MCP overview.

Next

Was this page helpful?

On this page