# Application Integration

> <!-- AUTO-GENERATED by export-skills.py — DO NOT EDIT -->

- Skill: `frank-luongt/application-integration` (Agent Skill)
- Install (CLI): `npx skillmds@latest add frank-luongt/application-integration`
- Raw SKILL.md: https://api.skillmd.com/api/skills/frank-luongt/application-integration/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Integrations & APIs
- Author: frank-luongt (https://skillmd.com/u/frank-luongt)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/frank-luongt/application-integration

---

<!-- AUTO-GENERATED by export-skills.py — DO NOT EDIT -->
---
name: application-integration
description: Google Cloud Application Integration and Integration Connectors for enterprise system connectivity. Use when connecting AI agents to SAP, Oracle, Salesforce, ServiceNow, Snowflake, or other enterprise systems via pre-built GCP connectors.
tags: [gcp, integration, connectors, enterprise]
---

# Google Cloud Application Integration

Connect AI agents to enterprise systems using Google Cloud's 100+ pre-built Integration Connectors for SAP, Oracle, Salesforce, ServiceNow, Snowflake, and more.

## When to Use

- Connecting Vertex AI agents to enterprise APIs without building custom middleware
- Using pre-built connectors for SAP (RFC/BAPI, OData), Oracle DB, Salesforce, ServiceNow
- Building event-driven AI pipelines triggered by enterprise system events
- Exposing enterprise data as Agent Builder OpenAPI tools

## Available Enterprise Connectors

| Category | Connectors |
|---|---|
| **ERP** | SAP ERP (RFC, BAPI), SAP S/4HANA (OData), Oracle ERP Cloud |
| **CRM** | Salesforce, HubSpot, Dynamics 365 |
| **ITSM** | ServiceNow |
| **Databases** | Oracle DB, MySQL, PostgreSQL, SQL Server, MongoDB, Cloud SQL, AlloyDB, Spanner |
| **Data Platforms** | Snowflake, BigQuery |
| **Messaging** | Pub/Sub, Kafka, RabbitMQ |
| **Storage** | Cloud Storage, S3, SFTP, FTP |
| **SaaS** | Zendesk, Workday, Slack |
| **Protocols** | REST, SOAP, gRPC, GraphQL |

## Patterns

### 1. SAP Connector (OData)

```yaml
# Integration connector configuration for SAP S/4HANA
connection:
  connector: sap
  connection_type: OData
  properties:
    sap_client: "100"
    odata_service_url: "https://sap-s4.example.com/sap/opu/odata/sap/API_SALES_ORDER_SRV"
    authentication:
      type: basic
      username_secret: "projects/my-project/secrets/sap-user/versions/latest"
      password_secret: "projects/my-project/secrets/sap-password/versions/latest"
```

### 2. Oracle DB Connector

```yaml
connection:
  connector: oracle_db
  properties:
    host: "oracle-db.example.com"
    port: 1521
    service_name: "ORCL"
    authentication:
      type: basic
      username_secret: "projects/my-project/secrets/oracle-user/versions/latest"
      password_secret: "projects/my-project/secrets/oracle-password/versions/latest"
```

### 3. Expose Connector as Agent Builder Tool

```python
# Create an OpenAPI spec wrapping Integration Connector for use as Agent Builder tool
openapi_spec = {
    "openapi": "3.0.0",
    "info": {"title": "SAP Sales Orders", "version": "1.0.0"},
    "servers": [{"url": "https://integrations.googleapis.com/v1/projects/MY_PROJECT/locations/us-central1/integrations/sap-sales-orders"}],
    "paths": {
        "/execute": {
            "post": {
                "operationId": "getSalesOrders",
                "summary": "Get SAP sales orders by customer",
                "requestBody": {
                    "content": {
                        "application/json": {
                            "schema": {
                                "type": "object",
                                "properties": {
                                    "customer_id": {"type": "string", "description": "SAP customer number"},
                                    "date_from": {"type": "string", "description": "Start date (YYYY-MM-DD)"},
                                },
                            }
                        }
                    }
                },
            }
        }
    },
}
```

### 4. Event-Driven AI Pipeline

```
SAP Sales Order Created
  --> Cloud Pub/Sub (via connector event trigger)
    --> Cloud Function
      --> Vertex AI Gemini (analyze order, check for anomalies)
        --> ServiceNow (create incident if anomaly detected, via connector)
```

## Anti-Patterns

- Building custom REST wrappers when a pre-built connector exists -- always check the connector catalog first
- Hardcoding credentials in connector configs -- use Secret Manager references
- Running high-frequency polling against SAP -- use event-driven triggers or batch sync
- Skipping connector testing -- verify connectivity before exposing to agents

## References

- [Application Integration Overview](https://cloud.google.com/application-integration/docs/overview)
- [Integration Connectors](https://cloud.google.com/integration-connectors/docs/overview)
- [Connector Reference](https://cloud.google.com/integration-connectors/docs/connectors)

<!-- Source: .faos/custom/skills/cloud/gcp/application-integration/SKILL.md -->