Integration Catalog
An integration is a reusable, credential-backed connection from a Conductor workflow to an external system — a model provider, database, queue, or third-party app.
Use this catalog when a workflow needs to call an external system — a model provider, a queue, a database, a cloud service, or a third-party app. Configure the integration once, then reuse it from workflows, AI tasks, system tasks, and remote services.
Sections
- AI / LLM — model providers (OpenAI, Anthropic, Bedrock, Vertex AI, Gemini, Mistral, Cohere, Grok, Perplexity, Hugging Face, Ollama) for chat, completion, and agent tasks.
- Vector Databases — Pinecone, Weaviate, pgvector, or MongoDB Atlas for embeddings, indexing, and retrieval.
- Message Broker — Kafka, Confluent, Amazon MSK, AMQP/RabbitMQ, NATS, SQS, Azure Service Bus, GCP Pub/Sub, or IBM MQ for event-driven orchestration.
- Cloud Providers — AWS or GCP for cloud storage and resource access.
- RDBMS — relational databases for querying and managing structured data.
- Email/Git — SendGrid for notification emails, Git for source-controlled files.
- Connected Apps — third-party productivity, project-management, and dev tools.