Quickstart Guide

Generate ISTQB-aligned test cases and execute them in a real browser, in under 5 minutes. No setup, no SDK, no programming required.

0Test Account Credentials

These credentials are for evaluation purposes only. A dedicated account with full access will be provisioned upon engagement.

1Direct API (curl)

Copy-paste these commands into any terminal (Mac/Linux/Windows PowerShell).

# Step 1: Get CSRF token
CSRF=$(curl -s -c cookies.txt \
  https://agent.goomni.ai/api/auth/csrf \
  | python3 -c "import sys,json; print(json.load(sys.stdin)['csrfToken'])")

# Step 2: Sign in (saves session to cookies.txt)
curl -s -b cookies.txt -c cookies.txt \
  -X POST https://agent.goomni.ai/api/auth/callback/credentials \
  -H "Content-Type: application/x-www-form-urlencoded" \
  -d "csrfToken=$CSRF&email=test-api@goomni.ai&password=GoomniTest2026!" \
  -o /dev/null -w "Auth status: %{http_code}\n"

# You should see: Auth status: 302 (success)

curl -s -b cookies.txt \
  -X POST https://agent.goomni.ai/api/qa/test-cases/generate \
  -H "Content-Type: application/json" \
  -d '{
    "requirements": "The system shall allow users to register with email and password. Password must be at least 8 characters with uppercase, lowercase, number, and special character. Failed login attempts are limited to 5 per 15-minute window.",
    "scope": "full",
    "maxTestCases": 15,
    "format": "standard"
  }' | python3 -m json.tool | head -80

# Create a user story
STORY=$(curl -s -b cookies.txt \
  -X POST https://agent.goomni.ai/api/pm/stories \
  -H "Content-Type: application/json" \
  -d '{
    "projectId": "550e8400-e29b-41d4-a716-446655440000",
    "organizationId": "660e8400-e29b-41d4-a716-446655440000",
    "title": "User login with MFA",
    "userStory": "As a user, I want to log in with multi-factor authentication, so that my account is secure",
    "priority": "critical",
    "acceptanceCriteria": [
      {"given": "a user with MFA enabled", "when": "they enter correct password", "then": "they receive an MFA code via SMS"},
      {"given": "a user enters wrong MFA code", "when": "3 attempts fail", "then": "the session is locked for 30 minutes"}
    ]
  }')
STORY_ID=$(echo $STORY | python3 -c "import sys,json; print(json.load(sys.stdin)['story']['storyId'])")
echo "Story created: $STORY_ID"

# AI-estimate story points
curl -s -b cookies.txt \
  -X POST "https://agent.goomni.ai/api/pm/stories/$STORY_ID/estimate" \
  -H "Content-Type: application/json" -d '{}' \
  | python3 -c "import sys,json; d=json.load(sys.stdin)['estimate']; print(f'Points: {d["points"]} | Complexity: {d["complexity"]} | Confidence: {d["confidence"]}')"

# Generate test cases bound to the story
curl -s -b cookies.txt \
  -X POST "https://agent.goomni.ai/api/pm/stories/$STORY_ID/generate-tests" \
  -H "Content-Type: application/json" \
  -d '{"format": "standard", "maxTestCases": 10}' \
  | python3 -c "import sys,json,re; d=json.loads(re.sub(r'[\x00-\x1f]',' ',sys.stdin.read())); [print(f'{tc["id"]}: {tc["title"]}') for tc in d['testCases']]"

# After generating test cases, export them:
curl -s -b cookies.txt \
  -X POST https://agent.goomni.ai/api/qa/test-cases/export \
  -H "Content-Type: application/json" \
  -d '{
    "testCases": [... paste test cases from previous step ...],
    "format": "csv",
    "projectName": "MyProject"
  }' -o test-cases.csv

# Supported formats: json, csv, gherkin, jira-xray, testrail, azure-devops

2Postman (Visual API Client)

Best for non-developers. Postman provides a visual interface, no terminal needed.

1. 1
   Download Postman from postman.com/downloads (free)
2. 2
   Import our collection: Open Postman, click Import, paste this URL:
   https://agent.goomni.ai/api/docs/postman
3. 3
   Set up authentication:

   1. In Postman, go to the collection's Authorization tab
   2. Select Bearer Token
   3. First run the CSRF and Sign In requests to get a session
   4. Postman auto-manages cookies after sign-in
4. 4
   Run "Generate Test Cases": Navigate to QA Test Cases > Generate, paste your requirements in the body, click Send.

3ChatGPT Integration

Use ChatGPT to interact with the API using natural language.

Option A: Custom GPT with Actions (Recommended)

1. 1
   Go to chat.openai.com/gpts/editor
2. 2
   Name: GoOmni QA Assistant
   Instructions:

   You are a QA test case generator. You help users create ISTQB-aligned test cases from their requirements using the GoOmni API. When a user provides requirements, call the test case generation endpoint. Always explain the generated test cases in plain language.

3. 3
   Click Create new action, paste the OpenAPI schema URL:
   https://agent.goomni.ai/api/docs/openapi
4. 4
   Set Authentication to API Key, header name: Cookie
   (Use the session cookie from Step 1a above)
5. 5
   Save and test: "Generate test cases for a user registration system with email, password, and social login"

Option B: Paste into ChatGPT directly (No setup)

I need you to generate test cases following ISTQB methodology for the following requirements:

[PASTE YOUR REQUIREMENTS HERE]

Use the GoOmni API to generate them. The API endpoint is:
POST https://agent.goomni.ai/api/qa/test-cases/generate

Request body:
{
  "requirements": "[the requirements text]",
  "scope": "full",
  "maxTestCases": 20,
  "format": "standard"
}

Please format the results as a table with columns: ID, Title, Priority, Test Type, Design Technique, Steps, Expected Result.

ChatGPT will format the output nicely even if you just describe your requirements in plain English.

4Claude (Anthropic) Integration

Connect Claude to GoOmni via MCP (Model Context Protocol) for native tool use.

Option A: Claude Desktop with MCP

Fastest path: open Settings → MCP access in GoOmni, generate an access key, and install the one-click GoOmni.mcpbbundle. To configure manually instead (requires Node.js):

1. 1
   Add to ~/Library/Application Support/Claude/claude_desktop_config.json. Claude Desktop only loads command-based servers, so bridge the endpoint with mcp-remote:

   {
     "mcpServers": {
       "goomni": {
         "command": "npx",
         "args": [
           "-y",
           "mcp-remote",
           "https://agent.goomni.ai/api/mcp",
           "--transport", "http-only",
           "--header", "Authorization: Bearer <your-jwt-token>"
         ]
       }
     }
   }

2. 2
   Fully quit and reopen Claude Desktop. You'll see GoOmni tools available in the tools menu.
3. 3
   Ask Claude: "Generate ISTQB test cases for a payment processing system that accepts Visa, Mastercard, and PayPal"

Option B: Claude API with tool use

import anthropic

client = anthropic.Anthropic()

# Define GoOmni as a tool
tools = [{
    "name": "generate_test_cases",
    "description": "Generate ISTQB-aligned test cases from requirements",
    "input_schema": {
        "type": "object",
        "properties": {
            "requirements": {"type": "string", "description": "Requirements text"},
            "maxTestCases": {"type": "integer", "default": 20}
        },
        "required": ["requirements"]
    }
}]

response = client.messages.create(
    model="claude-sonnet-4-6",
    max_tokens=4096,
    tools=tools,
    messages=[{
        "role": "user",
        "content": "Generate test cases for: User registration with email/password, password must be 8+ chars with uppercase, lowercase, number, and special char."
    }]
)

# When Claude calls the tool, forward to GoOmni API
# POST https://agent.goomni.ai/api/qa/test-cases/generate

5Full Lifecycle (End-to-End)

The complete flow from requirements to documented, tested software.