exasol-extension-catalog

name	exasol-extension-catalog
description	Catalog and routing guide for Exasol tools, extensions, connectors, integrations, and architecture patterns. Use when users ask how to extend, customize, deploy, load, explore, enrich, surface, scale, automate, govern, migrate, or integrate Exasol, including Exasol Labs tools, AI/ML, UDFs, lakehouse, SAP, Databricks, agentic workflows, or product/version status.

Exasol Extension Catalog Skill

Last reviewed: 2026-07-03.

Use this skill when a user asks how to extend, customize, integrate, deploy, operate, automate, enrich, expose, or scale Exasol.

The catalog organizes Exasol capabilities into six categories:

• DEPLOY: provision, install, package, configure, schedule, and operate Exasol.
• LOAD: ingest, import, federate, stream, move, or query external data.
• EXPLORE: inspect schemas, query data, use notebooks, BI tools, catalogs, and agentic query interfaces.
• ENRICH: add AI, ML, UDFs, text processing, semantic interpretation, agents, and custom computation.
• SURFACE: expose Exasol to humans, applications, BI, APIs, low-code tools, and agents.
• SCALE: improve scale-out, performance, reliability, governance, observability, testing, repeatability, and enterprise operations.

Important Use Rules

1. Prefer official Exasol docs or GitHub repositories when giving installation or configuration instructions.
2. Distinguish between Exasol-owned or Exasol-maintained tools, Exasol Labs/community tools, and third-party ecosystem integrations.
3. For customer-facing recommendations, state support level when known. If support level is not explicit in the reference, verify the linked source before presenting it as supported.
4. For current versions, latest releases, security status, or support status, check the linked source before answering.
5. For destructive operations, deployment changes, credential management, RBAC, Terraform, or agent automation, include safety and rollback guidance.
6. For agentic workflows, prefer governed patterns: MCP Server, Agent Control Plane, semantic layer, SQL guardrails, audit logs, and least-privilege credentials.

Routing Algorithm

Choose the narrowest matching category and load only the matching reference files. If multiple categories apply, load all relevant references before answering.

1. Run Exasol somewhere or operate deployment tooling

   • Trigger phrases: deploy, install, Exasol SaaS, Exasol Personal, Docker DB, Terraform, Ansible, scheduler, public keys, artifact verification, Extension Manager
   • Load: references/deploy.md

2. Bring data into Exasol or query data elsewhere

   • Trigger phrases: load, ingest, import, federate, Virtual Schema, Databricks, SAP HANA, Snowflake, Oracle, PostgreSQL, Kafka, Kinesis, Spark, Glue, Azure Data Factory, migration, driver, ODBC, JDBC, PyExasol
   • Load: references/load.md

3. Understand data or query it interactively

   • Trigger phrases: explore, MCP, Text-to-SQL, VS Code, AI Lab, Notebook Connector, SQL client, catalog, semantic layer, metadata
   • Load: references/explore.md

4. Add AI, ML, UDFs, text analytics, or agents

   • Trigger phrases: AI, ML, transformers, sentiment, entity extraction, SageMaker, MLflow, model UDF, foundation model, Agent Control Plane, Rust UDF, Metadata Agent
   • Load: references/enrich.md

5. Expose Exasol to apps, BI, APIs, users, or agents

   • Trigger phrases: BI, dashboard, Tableau, Power BI, Superset, REST API, ERA, Power Apps, low-code, application driver, CData, Denodo, Trino
   • Load: references/surface.md

6. Improve scale, governance, performance, observability, repeatability, or reliability

   • Trigger phrases: scale, performance, governance, RBAC, observability, CloudWatch, telemetry, SBOM, testing, CI, change management, data protection, warehouse automation
   • Load: references/scale.md

Fast Routing Guide

• User wants to run Exasol somewhere: use DEPLOY.
• User wants to bring data into Exasol or query data elsewhere: use LOAD.
• User wants to understand data or query it interactively: use EXPLORE.
• User wants AI, ML, UDFs, text analytics, or agents: use ENRICH.
• User wants to expose Exasol to apps, BI, APIs, users, or agents: use SURFACE.
• User wants production scale, governance, performance, observability, repeatability, or reliability: use SCALE.

Related Skills

• Use exasol-database for direct SQL, exapump, table design, and query profiling after choosing the relevant tool family.
• Use exasol-import for direct native IMPORT, local CSV or Parquet loading, connection objects for import, and native Parquet loading behavior.
• Use exasol-export for direct native EXPORT, local file export, and connection objects for export.
• Use exasol-cloud-storage-extension for Cloud Storage Extension import/export workflows after choosing that extension family.
• Use exasol-jdbc-virtual-schemas for JDBC/database-source virtual schema usage after choosing that federation family.
• Use exasol-document-virtual-schemas for document-file virtual schema usage (S3, GCS, Azure object storage) after choosing that federation family.
• Use exasol-virtual-schema-adapter-development for custom virtual schema adapter implementation, packaging, and adapter-side debugging after choosing that a maintained adapter is not enough.
• Use exasol-bucketfs for BucketFS file upload, download, list, and delete workflows.
• Use exasol-udfs for UDF and Script Language Container implementation details.
• Use exasol-ai-setup and exasol-notebook-connections for notebook-connector configuration and Python connection helpers.
• Use exasol-setup-personal for guided Exasol Personal deployment.