Middleware Testing Patterns logo

Middleware Testing Patterns

Community
proffesor-for-testing
middleware-testing-patterns

Enterprise middleware testing patterns for message routing, transformation, DLQ, protocol mediation, ESB error handling, and EIP patterns. Use when testing middleware layers, message brokers, ESBs, or integration buses.

Overview

Publisherproffesor-for-testing
Repositoryagentic-qe
Skill namemiddleware-testing-patterns
Stars
480
Forks
92
Bundled files
3
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by proffesor-for-testing on GitHub. Read the source before you install it.

Installation

Install the Middleware Testing Patterns AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/proffesor-for-testing/agentic-qe.git /tmp/agentic-qe
mkdir -p .claude/skills
cp -r /tmp/agentic-qe/assets/skills/middleware-testing-patterns .claude/skills/middleware-testing-patterns
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Middleware Testing Patterns in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Middleware Testing Patterns on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Middleware Testing Patterns is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Middleware Testing Patterns

<default_to_action> When testing enterprise middleware, ESBs, or message-driven systems:

  1. VALIDATE message routing logic (content-based, header-based, recipient list)
  2. TEST message transformations end-to-end (input format -> output format)
  3. VERIFY dead letter queue handling (poison messages, retry exhaustion)
  4. ASSERT message ordering and sequencing with correlation IDs
  5. EXERCISE error handling and compensation patterns
  6. TEST protocol mediation (SOAP-to-REST, sync-to-async)
  7. VALIDATE EIP patterns (splitter, aggregator, content enricher, normalizer)

Quick Pattern Selection:

  • Message routing issues -> Content-based router tests
  • Transformation failures -> Schema-in/schema-out validation
  • Lost messages -> DLQ and retry pattern tests
  • Protocol bridging -> Mediation round-trip tests
  • Complex flows -> Correlation ID tracing tests

Critical Success Factors:

  • Middleware is invisible when it works; test the invisible
  • Always validate both the happy path AND the error channel (DLQ)
  • Correlation IDs are your lifeline for tracing multi-hop messages </default_to_action>

Quick Reference Card

When to Use

  • Testing ESB message flows (IBM IIB/ACE, MuleSoft, WSO2)
  • Validating message transformations (XSLT, JSON-to-XML, flat-file parsing)
  • Testing message broker routing (MQ, Kafka, RabbitMQ)
  • Verifying dead letter queue behavior
  • Testing protocol mediation (SOAP/REST bridging)
  • Validating Enterprise Integration Patterns (EIP)

Testing Levels

LevelPurposeDependenciesSpeed
Unit TransformSingle mapping correctnessNoneFast
Route LogicRouting decision accuracyMocked endpointsFast
IntegrationEnd-to-end message flowBroker + endpointsMedium
DLQ/ErrorError handling and recoveryFull middleware stackSlower

Critical Test Scenarios

ScenarioMust TestExample
RoutingCorrect destination selectionOrder type A -> Queue A, type B -> Queue B
TransformationSchema compliance after mappingXML -> JSON field mapping accuracy
DLQPoison message handlingMalformed XML lands in DLQ, not lost
OrderingSequence preservationMessages 1-2-3 arrive in order
CorrelationMulti-hop tracingRequest-reply matched by correlation ID
RetryTransient failure recovery3 retries with backoff, then DLQ
MediationProtocol bridging fidelitySOAP request produces correct REST call

Tools

  • Message Brokers: IBM MQ, RabbitMQ, Apache Kafka, ActiveMQ
  • ESBs: IBM IIB/ACE, MuleSoft, WSO2, Apache Camel
  • Testing: SoapUI, Postman, custom harnesses
  • Virtualization: Mountebank, WireMock, HoverFly
  • Monitoring: Splunk, ELK, Datadog

Agent Coordination

  • qe-middleware-validator: Validate routing rules, transformation accuracy, EIP patterns
  • qe-message-broker-tester: Test broker connectivity, DLQ behavior, message ordering
  • qe-soap-tester: SOAP/WSDL validation, WS-Security, protocol mediation

Message Routing Pattern Testing

Content-Based Router

javascript
describe('Content-Based Router - Order Type', () => {
  it('routes standard orders to fulfillment queue', async () => {
    const message = {
      correlationId: uuid(),
      body: { orderType: 'STANDARD', orderId: 'ORD-001', items: [{ sku: 'A1', qty: 2 }] }
    };

    await broker.publish('orders.inbound', message);

    const routed = await broker.consume('orders.fulfillment', { timeout: 5000 });
    expect(routed.body.orderId).toBe('ORD-001');
    expect(routed.correlationId).toBe(message.correlationId);

    const dlq = await broker.tryConsume('orders.dlq', { timeout: 1000 });
    expect(dlq).toBeNull(); // Nothing in DLQ
  });

  it('routes express orders to priority queue', async () => {
    const message = {
      correlationId: uuid(),
      body: { orderType: 'EXPRESS', orderId: 'ORD-002', items: [{ sku: 'B1', qty: 1 }] }
    };

    await broker.publish('orders.inbound', message);

    const routed = await broker.consume('orders.priority', { timeout: 5000 });
    expect(routed.body.orderId).toBe('ORD-002');
  });

  it('sends unrecognized order types to DLQ', async () => {
    const message = {
      correlationId: uuid(),
      body: { orderType: 'UNKNOWN', orderId: 'ORD-003' }
    };

    await broker.publish('orders.inbound', message);

    const dlqMessage = await broker.consume('orders.dlq', { timeout: 5000 });
    expect(dlqMessage.body.orderId).toBe('ORD-003');
    expect(dlqMessage.headers['x-error-reason']).toContain('Unrecognized orderType');
  });
});

Header-Based Router

javascript
describe('Header-Based Router - Region', () => {
  it('routes by x-region header to regional queues', async () => {
    const regions = ['US', 'EU', 'APAC'];

    for (const region of regions) {
      const message = {
        headers: { 'x-region': region },
        body: { orderId: `ORD-${region}` }
      };

      await broker.publish('orders.global', message);

      const routed = await broker.consume(`orders.${region.toLowerCase()}`, { timeout: 5000 });
      expect(routed.body.orderId).toBe(`ORD-${region}`);
    }
  });

  it('routes messages without region header to default queue', async () => {
    await broker.publish('orders.global', { body: { orderId: 'ORD-NO-REGION' } });

    const routed = await broker.consume('orders.default', { timeout: 5000 });
    expect(routed.body.orderId).toBe('ORD-NO-REGION');
  });
});

Recipient List

javascript
describe('Recipient List - Order Notifications', () => {
  it('fans out order to all registered recipients', async () => {
    const message = {
      body: { orderId: 'ORD-100', total: 250.00 }
    };

    await broker.publish('orders.confirmed', message);

    // All three systems should receive the message
    const [warehouse, billing, crm] = await Promise.all([
      broker.consume('notify.warehouse', { timeout: 5000 }),
      broker.consume('notify.billing', { timeout: 5000 }),
      broker.consume('notify.crm', { timeout: 5000 })
    ]);

    expect(warehouse.body.orderId).toBe('ORD-100');
    expect(billing.body.orderId).toBe('ORD-100');
    expect(crm.body.orderId).toBe('ORD-100');
  });
});

Message Transformation Testing

JSON-to-XML Transformation

javascript
describe('JSON to XML Transformation', () => {
  it('transforms order JSON to SAP XML format', async () => {
    const jsonInput = {
      orderId: 'ORD-500',
      customer: { id: 'CUST-01', name: 'Acme Corp' },
      items: [
        { sku: 'MAT-100', quantity: 10, unitPrice: 25.00 }
      ]
    };

    await broker.publish('transform.json-in', { body: jsonInput });

    const xmlOutput = await broker.consume('transform.xml-out', { timeout: 5000 });
    const parsed = parseXml(xmlOutput.body);

    expect(parsed.SalesOrder.OrderNumber).toBe('ORD-500');
    expect(parsed.SalesOrder.SoldToParty.CustomerID).toBe('CUST-01');
    expect(parsed.SalesOrder.Items.Item[0].MaterialNumber).toBe('MAT-100');
    expect(parsed.SalesOrder.Items.Item[0].Quantity).toBe('10');
  });

  it('handles missing optional fields gracefully', async () => {
    const jsonInput = {
      orderId: 'ORD-501',
      customer: { id: 'CUST-02' }, // name is missing
      items: [{ sku: 'MAT-200', quantity: 5 }] // unitPrice missing
    };

    await broker.publish('transform.json-in', { body: jsonInput });

    const xmlOutput = await broker.consume('transform.xml-out', { timeout: 5000 });
    const parsed = parseXml(xmlOutput.body);

    expect(parsed.SalesOrder.SoldToParty.Name).toBe(''); // Default empty
    expect(parsed.SalesOrder.Items.Item[0].UnitPrice).toBe('0.00');
  });

  it('rejects invalid input and sends to DLQ with error details', async () => {
    const invalidInput = { items: [] }; // Missing required orderId and customer

    await broker.publish('transform.json-in', { body: invalidInput });

    const dlq = await broker.consume('transform.dlq', { timeout: 5000 });
    expect(dlq.headers['x-error-reason']).toContain('orderId is required');
  });
});

Flat-File to XML Transformation

javascript
describe('Flat-File to XML Transformation', () => {
  it('parses fixed-width flat file into structured XML', async () => {
    // Fixed-width: positions 1-10 = orderId, 11-20 = sku, 21-25 = qty
    const flatFile = 'ORD-001   MAT-10000010\nORD-001   MAT-20000005\n';

    await broker.publish('transform.flatfile-in', { body: flatFile });

    const xmlOutput = await broker.consume('transform.flatfile-xml-out', { timeout: 5000 });
    const parsed = parseXml(xmlOutput.body);

    expect(parsed.Orders.Order).toHaveLength(1);
    expect(parsed.Orders.Order[0].Items.Item).toHaveLength(2);
    expect(parsed.Orders.Order[0].Items.Item[0].Quantity).toBe('10');
  });

  it('handles lines with trailing whitespace correctly', async () => {
    const flatFile = 'ORD-002   MAT-30000001   \n';

    await broker.publish('transform.flatfile-in', { body: flatFile });

    const xmlOutput = await broker.consume('transform.flatfile-xml-out', { timeout: 5000 });
    const parsed = parseXml(xmlOutput.body);

    expect(parsed.Orders.Order[0].OrderId).toBe('ORD-002');
    expect(parsed.Orders.Order[0].Items.Item[0].Quantity).toBe('1');
  });
});

Dead Letter Queue (DLQ) Testing

javascript
describe('Dead Letter Queue Strategy', () => {
  it('moves message to DLQ after max retries', async () => {
    // Configure endpoint to always fail
    mockEndpoint.respondWith(500, 'Service Unavailable');

    await broker.publish('orders.process', {
      body: { orderId: 'ORD-FAIL' },
      headers: { 'x-correlation-id': 'corr-123' }
    });

    // Wait for retries to exhaust (3 retries with backoff)
    const dlqMessage = await broker.consume('orders.dlq', { timeout: 30000 });

    expect(dlqMessage.body.orderId).toBe('ORD-FAIL');
    expect(dlqMessage.headers['x-retry-count']).toBe('3');
    expect(dlqMessage.headers['x-original-queue']).toBe('orders.process');
    expect(dlqMessage.headers['x-correlation-id']).toBe('corr-123');
    expect(dlqMessage.headers['x-error-reason']).toContain('Service Unavailable');
    expect(dlqMessage.headers['x-first-failure-timestamp']).toBeDefined();
  });

  it('preserves original message body in DLQ', async () => {
    const originalBody = {
      orderId: 'ORD-PRESERVE',
      items: [{ sku: 'A', qty: 1 }, { sku: 'B', qty: 2 }]
    };

    mockEndpoint.respondWith(500, 'Timeout');
    await broker.publish('orders.process', { body: originalBody });

    const dlqMessage = await broker.consume('orders.dlq', { timeout: 30000 });
    expect(dlqMessage.body).toEqual(originalBody); // Body unchanged
  });

  it('DLQ messages can be replayed after fix', async () => {
    // Step 1: Message fails, lands in DLQ
    mockEndpoint.respondWith(500, 'Down');
    await broker.publish('orders.process', { body: { orderId: 'ORD-REPLAY' } });
    const dlqMessage = await broker.consume('orders.dlq', { timeout: 30000 });

    // Step 2: Fix the endpoint
    mockEndpoint.respondWith(200, 'OK');

    // Step 3: Replay from DLQ
    await broker.publish('orders.process', dlqMessage);
    const success = await broker.consume('orders.completed', { timeout: 5000 });
    expect(success.body.orderId).toBe('ORD-REPLAY');
  });
});

Message Ordering and Sequencing

javascript
describe('Message Ordering Validation', () => {
  it('preserves FIFO order within same partition key', async () => {
    const messages = Array.from({ length: 10 }, (_, i) => ({
      headers: { 'x-partition-key': 'ORDER-GROUP-1', 'x-sequence': String(i + 1) },
      body: { orderId: `ORD-${i + 1}`, sequence: i + 1 }
    }));

    for (const msg of messages) {
      await broker.publish('orders.sequenced', msg);
    }

    const received = [];
    for (let i = 0; i < 10; i++) {
      const msg = await broker.consume('orders.sequenced.out', { timeout: 5000 });
      received.push(msg.body.sequence);
    }

    expect(received).toEqual([1, 2, 3, 4, 5, 6, 7, 8, 9, 10]);
  });

  it('detects out-of-order messages via sequence gap', async () => {
    // Send messages 1, 2, 4 (skip 3)
    for (const seq of [1, 2, 4]) {
      await broker.publish('orders.sequenced', {
        headers: { 'x-partition-key': 'GROUP-2', 'x-sequence': String(seq) },
        body: { sequence: seq }
      });
    }

    // The middleware should detect the gap and raise an alert
    const alert = await broker.consume('orders.sequence-gap', { timeout: 10000 });
    expect(alert.body.missingSequence).toBe(3);
    expect(alert.body.partitionKey).toBe('GROUP-2');
  });
});

ESB Error Handling and Compensation

javascript
describe('Compensation Pattern - Saga', () => {
  it('compensates completed steps when a later step fails', async () => {
    // Saga: Reserve Inventory -> Charge Payment -> Ship
    // If Ship fails, Payment and Inventory must be compensated

    mockServices.inventory.respondWith(200); // Step 1 succeeds
    mockServices.payment.respondWith(200);   // Step 2 succeeds
    mockServices.shipping.respondWith(500);  // Step 3 fails

    await broker.publish('saga.order', {
      body: { orderId: 'ORD-SAGA', amount: 100.00 }
    });

    // Verify compensation messages were sent
    const paymentRefund = await broker.consume('saga.compensate.payment', { timeout: 10000 });
    expect(paymentRefund.body.action).toBe('REFUND');
    expect(paymentRefund.body.orderId).toBe('ORD-SAGA');

    const inventoryRelease = await broker.consume('saga.compensate.inventory', { timeout: 10000 });
    expect(inventoryRelease.body.action).toBe('RELEASE');
    expect(inventoryRelease.body.orderId).toBe('ORD-SAGA');
  });
});

Protocol Mediation Testing

SOAP-to-REST Mediation

javascript
describe('SOAP to REST Protocol Mediation', () => {
  it('converts SOAP request to REST call and back', async () => {
    const soapRequest = `
      <soapenv:Envelope xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/"
                        xmlns:ord="http://example.com/orders">
        <soapenv:Body>
          <ord:GetOrder>
            <ord:OrderId>ORD-700</ord:OrderId>
          </ord:GetOrder>
        </soapenv:Body>
      </soapenv:Envelope>
    `;

    // ESB should convert SOAP -> REST GET /orders/ORD-700
    mockRestBackend.onGet('/orders/ORD-700').reply(200, {
      orderId: 'ORD-700',
      status: 'SHIPPED',
      total: 150.00
    });

    const response = await soapClient.send(soapRequest, { endpoint: esbSoapEndpoint });

    // Verify SOAP response wraps REST data correctly
    const parsed = parseSoapResponse(response);
    expect(parsed.OrderId).toBe('ORD-700');
    expect(parsed.Status).toBe('SHIPPED');
    expect(parsed.Total).toBe('150.00');
  });

  it('maps REST error to SOAP fault', async () => {
    const soapRequest = buildSoapGetOrder('ORD-NOTFOUND');

    mockRestBackend.onGet('/orders/ORD-NOTFOUND').reply(404, {
      error: 'Order not found'
    });

    const response = await soapClient.send(soapRequest, { endpoint: esbSoapEndpoint });
    const fault = parseSoapFault(response);

    expect(fault.faultCode).toBe('Client');
    expect(fault.faultString).toContain('Order not found');
  });
});

Sync-to-Async Mediation

javascript
describe('Sync to Async Mediation', () => {
  it('accepts sync request, returns correlation ID, delivers result async', async () => {
    // Client sends synchronous HTTP POST
    const response = await httpClient.post(`${esbEndpoint}/async-orders`, {
      orderId: 'ORD-ASYNC',
      items: [{ sku: 'X1', qty: 5 }]
    });

    // ESB returns 202 Accepted with correlation ID
    expect(response.status).toBe(202);
    expect(response.body.correlationId).toBeDefined();
    const correlationId = response.body.correlationId;

    // Result arrives asynchronously on callback queue
    const result = await broker.consume('orders.async.results', {
      timeout: 15000,
      filter: (msg) => msg.headers['x-correlation-id'] === correlationId
    });

    expect(result.body.orderId).toBe('ORD-ASYNC');
    expect(result.body.status).toBe('ACCEPTED');
  });
});

Enterprise Integration Patterns (EIP)

Splitter Pattern

javascript
describe('Splitter - Batch Order to Individual Items', () => {
  it('splits batch order into individual item messages', async () => {
    const batchOrder = {
      orderId: 'BATCH-001',
      items: [
        { sku: 'A1', qty: 10 },
        { sku: 'B2', qty: 5 },
        { sku: 'C3', qty: 20 }
      ]
    };

    await broker.publish('orders.batch', { body: batchOrder });

    const splitMessages = [];
    for (let i = 0; i < 3; i++) {
      splitMessages.push(await broker.consume('orders.items', { timeout: 5000 }));
    }

    expect(splitMessages).toHaveLength(3);
    expect(splitMessages.map(m => m.body.sku).sort()).toEqual(['A1', 'B2', 'C3']);

    // Each split message retains parent correlation
    for (const msg of splitMessages) {
      expect(msg.headers['x-batch-id']).toBe('BATCH-001');
      expect(msg.headers['x-split-total']).toBe('3');
    }
  });
});

Aggregator Pattern

javascript
describe('Aggregator - Collect Item Results', () => {
  it('aggregates individual item results into batch response', async () => {
    const batchId = 'BATCH-002';
    const itemResults = [
      { sku: 'A1', status: 'PICKED', batchId },
      { sku: 'B2', status: 'PICKED', batchId },
      { sku: 'C3', status: 'PICKED', batchId }
    ];

    for (const result of itemResults) {
      await broker.publish('items.completed', {
        headers: { 'x-batch-id': batchId, 'x-split-total': '3' },
        body: result
      });
    }

    const aggregated = await broker.consume('orders.batch.completed', { timeout: 10000 });

    expect(aggregated.body.batchId).toBe('BATCH-002');
    expect(aggregated.body.items).toHaveLength(3);
    expect(aggregated.body.items.every(i => i.status === 'PICKED')).toBe(true);
  });

  it('handles aggregation timeout for incomplete batches', async () => {
    const batchId = 'BATCH-TIMEOUT';

    // Only send 2 of 3 expected items
    for (let i = 0; i < 2; i++) {
      await broker.publish('items.completed', {
        headers: { 'x-batch-id': batchId, 'x-split-total': '3' },
        body: { sku: `ITEM-${i}`, status: 'PICKED', batchId }
      });
    }

    // After timeout, aggregator should emit partial result with error
    const partial = await broker.consume('orders.batch.incomplete', { timeout: 60000 });
    expect(partial.body.batchId).toBe('BATCH-TIMEOUT');
    expect(partial.body.receivedCount).toBe(2);
    expect(partial.body.expectedCount).toBe(3);
  });
});

Content Enricher Pattern

javascript
describe('Content Enricher - Customer Data Lookup', () => {
  it('enriches order with customer details from CRM', async () => {
    mockCrm.onGet('/customers/CUST-50').reply(200, {
      name: 'Acme Corp',
      tier: 'GOLD',
      creditLimit: 50000
    });

    await broker.publish('orders.enrich', {
      body: { orderId: 'ORD-ENRICH', customerId: 'CUST-50', total: 1200 }
    });

    const enriched = await broker.consume('orders.enriched', { timeout: 5000 });

    expect(enriched.body.orderId).toBe('ORD-ENRICH');
    expect(enriched.body.customer.name).toBe('Acme Corp');
    expect(enriched.body.customer.tier).toBe('GOLD');
    expect(enriched.body.creditCheckPassed).toBe(true); // 1200 < 50000
  });
});

Retry and Circuit Breaker Patterns

javascript
describe('Retry with Exponential Backoff', () => {
  it('retries 3 times with increasing delays before DLQ', async () => {
    const timestamps = [];
    mockEndpoint.onPost('/process').reply(() => {
      timestamps.push(Date.now());
      return [500, 'Temporary failure'];
    });

    await broker.publish('orders.process', { body: { orderId: 'ORD-RETRY' } });

    await broker.consume('orders.dlq', { timeout: 60000 });

    // Verify exponential backoff timing (1s, 2s, 4s approximately)
    expect(timestamps).toHaveLength(4); // 1 initial + 3 retries
    const delay1 = timestamps[1] - timestamps[0];
    const delay2 = timestamps[2] - timestamps[1];
    expect(delay1).toBeGreaterThan(800);  // ~1s
    expect(delay2).toBeGreaterThan(1800); // ~2s
  });
});

describe('Circuit Breaker', () => {
  it('opens circuit after threshold failures', async () => {
    // Fail 5 messages to trip circuit breaker (threshold = 5)
    mockEndpoint.onPost('/process').reply(500, 'Down');

    for (let i = 0; i < 5; i++) {
      await broker.publish('orders.circuit', { body: { orderId: `ORD-CB-${i}` } });
    }

    // Wait for circuit to open
    await sleep(5000);

    // Next message should be immediately rejected (circuit open)
    const startTime = Date.now();
    await broker.publish('orders.circuit', { body: { orderId: 'ORD-CB-FAST-FAIL' } });
    const dlq = await broker.consume('orders.circuit.dlq', { timeout: 3000 });
    const elapsed = Date.now() - startTime;

    expect(dlq.headers['x-circuit-state']).toBe('OPEN');
    expect(elapsed).toBeLessThan(2000); // Fast fail, no retry
  });
});

Service Virtualization for Isolated Testing

javascript
describe('Service Virtualization - Middleware Isolation', () => {
  let virtualService;

  beforeAll(async () => {
    virtualService = await mountebank.createImposter({
      port: 4545,
      protocol: 'http',
      stubs: [
        {
          predicates: [{ equals: { method: 'POST', path: '/sap/orders' } }],
          responses: [{
            is: {
              statusCode: 200,
              headers: { 'Content-Type': 'application/xml' },
              body: '<OrderResponse><Status>CREATED</Status><SapOrderId>SAP-001</SapOrderId></OrderResponse>'
            }
          }]
        },
        {
          predicates: [{ equals: { method: 'POST', path: '/sap/orders' },
                         contains: { body: '<InvalidField>' } }],
          responses: [{
            is: { statusCode: 400, body: '<Error>Invalid request</Error>' }
          }]
        }
      ]
    });
  });

  afterAll(async () => {
    await mountebank.deleteImposter(4545);
  });

  it('middleware correctly handles SAP success response', async () => {
    await broker.publish('orders.to-sap', {
      body: { orderId: 'ORD-VIRT', customer: 'CUST-01' }
    });

    const result = await broker.consume('orders.sap-confirmed', { timeout: 5000 });
    expect(result.body.sapOrderId).toBe('SAP-001');
  });
});

Best Practices

Do This

  • Test both the happy path AND the error channel (DLQ) for every flow
  • Use correlation IDs to trace messages across multi-hop middleware
  • Validate schema compliance at every transformation boundary
  • Test message ordering with partitioned/sequenced queues
  • Use service virtualization to isolate middleware from backend systems
  • Test retry exhaustion to confirm DLQ behavior
  • Validate compensation patterns (saga rollback) for multi-step flows

Avoid This

  • Testing only the happy path and ignoring DLQ/error queues
  • Hardcoding queue names; use configuration-driven tests
  • Ignoring message ordering requirements
  • Skipping transformation edge cases (empty fields, special characters, encoding)
  • Testing middleware without correlation ID validation
  • Assuming retries work correctly without verifying backoff timing
  • Deploying without testing the circuit breaker threshold

Agent-Assisted Middleware Testing

typescript
// Validate middleware routing rules
await Task("Middleware Routing Validation", {
  routingConfig: 'esb/routing-rules.xml',
  messageTypes: ['STANDARD_ORDER', 'EXPRESS_ORDER', 'RETURN'],
  validateDLQ: true,
  checkCorrelationIds: true
}, "qe-middleware-validator");

// Test message broker behavior
await Task("Message Broker DLQ Test", {
  broker: 'rabbitmq',
  queues: ['orders.inbound', 'orders.process', 'orders.dlq'],
  retryPolicy: { maxRetries: 3, backoff: 'exponential' },
  poisonMessages: ['malformed-xml', 'missing-required-fields', 'oversized-payload']
}, "qe-message-broker-tester");

// SOAP/WSDL contract validation
await Task("SOAP Service Contract Validation", {
  wsdl: 'services/OrderService.wsdl',
  endpoints: ['CreateOrder', 'GetOrderStatus', 'CancelOrder'],
  validateWsSecurity: true,
  testFaultHandling: true
}, "qe-soap-tester");

// Transformation accuracy testing
await Task("Transformation Accuracy Suite", {
  transformations: [
    { name: 'JSON-to-SAP-XML', input: 'testdata/order.json', expected: 'testdata/order-sap.xml' },
    { name: 'Flat-to-XML', input: 'testdata/inventory.dat', expected: 'testdata/inventory.xml' }
  ],
  validateSchemas: true,
  edgeCases: ['empty-fields', 'special-chars', 'max-length']
}, "qe-middleware-validator");

Agent Coordination Hints

Memory Namespace

aqe/middleware-testing/
  routing/           - Routing rule configurations and test results
  transformations/   - Mapping definitions and validation results
  dlq/               - DLQ test results and replay outcomes
  eip-patterns/      - EIP pattern test coverage
  circuit-breakers/  - Circuit breaker state and threshold configs
  correlation/       - Correlation ID tracing results

Fleet Coordination

typescript
const middlewareFleet = await FleetManager.coordinate({
  strategy: 'middleware-testing',
  agents: [
    'qe-middleware-validator',   // Routing, transformation, EIP patterns
    'qe-message-broker-tester',  // DLQ, ordering, retry, circuit breaker
    'qe-soap-tester'             // SOAP contracts, WS-Security, mediation
  ],
  topology: 'mesh'
});

await middlewareFleet.execute({
  flows: [
    { name: 'order-inbound', stages: ['route', 'transform', 'enrich', 'deliver'] },
    { name: 'payment-processing', stages: ['validate', 'authorize', 'settle'] }
  ],
  testDLQ: true,
  testCompensation: true
});

Related Skills


Remember

Middleware is the nervous system of enterprise architecture. When it works, nobody notices. When it fails, everything breaks. Test the routing decisions, the transformations, the error channels, and the compensation patterns. Always validate with correlation IDs across multi-hop flows, and always confirm that poison messages end up in the DLQ with enough metadata to diagnose and replay.

With Agents: Agents validate routing rules against message types, test transformation accuracy with schema compliance, and exercise DLQ/retry/circuit-breaker behavior systematically. Use agents to cover the combinatorial explosion of routing paths and transformation edge cases that manual testing cannot reach.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Middleware Testing Patterns AI skill do?

Enterprise middleware testing patterns for message routing, transformation, DLQ, protocol mediation, ESB error handling, and EIP patterns. Use when testing middleware layers, message brokers, ESBs, or integration buses.

Why use Middleware Testing Patterns on TypingMind?

Because you install it once and use it with any model. Middleware Testing Patterns is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Middleware Testing Patterns in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/proffesor-for-testing/agentic-qe/tree/main/assets/skills/middleware-testing-patterns. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Middleware Testing Patterns?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Middleware Testing Patterns?

As many as you like. As long as a model supports skills, you can use Middleware Testing Patterns with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Middleware Testing Patterns AI skill free?

Yes. It is published on GitHub by proffesor-for-testing under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇