Exasol Database Skill

Trigger when the user mentions Exasol, exapump, database import/export, CSV/Parquet upload, Exasol SQL, IMPORT INTO, EXPORT INTO, EXA_, or any Exasol database interaction.

Step 0: Establish Connection

Ensure a working exapump profile before proceeding:

• If the user mentions a specific profile name → test it: exapump sql --profile <name> "SELECT 1" (always place --profile after the subcommand). On success, use --profile <name> on all subsequent commands.

• Otherwise → test the default profile: exapump sql "SELECT 1" .

• On success → proceed. No further connection setup needed.

• On failure → run exapump profile list to check available profiles.

• If profiles exist → present the list and ask the user which to use, then retry with exapump sql --profile <name> "SELECT 1" (always place --profile after the subcommand).

• If no profiles → tell the user to run exapump profile add default to create one, then retry.

• Never read or reference the exapump configuration file — it contains credentials.

Routing Algorithm

After the connection is established, determine the task type and load only the references needed:

Local files (upload CSV/Parquet, export to local files):

• Load: references/exapump-reference.md

• Load: references/import-export.md (decision tree, connection objects)

Remote files / bulk loading (S3, Azure, GCS, FTP, HTTP — IMPORT/EXPORT):

• Load: references/import-export.md

SQL execution (queries, DDL, DML, schema inspection):

• Load: references/exapump-reference.md (CLI usage)

• Load: references/exasol-sql.md (core SQL behavior)

Table design (DISTRIBUTE BY, PARTITION BY, CREATE TABLE layout):

• Load: references/table-design.md

Query profiling / performance (slow queries, data skew, REORGANIZE):

• Load: references/query-profiling.md

Analytics / window functions (ROW_NUMBER, RANK, LAG/LEAD, QUALIFY, GROUPING SETS):

• Load: references/analytics-qualify.md

Virtual Schemas (external data sources, adapter scripts):

• Load: references/virtual-schemas.md

Multiple routes can apply — load all that match.

• Before writing any SQL (applies to routes 2–7):

• Always double-quote every identifier (column names, table names, schema names) in SELECT, FROM, WHERE, GROUP BY, ORDER BY, and JOIN clauses — without exception

• This preserves mixed-case names and prevents reserved-keyword errors in a single rule

• Do NOT quote SQL keywords, functions, or aliases — only object identifiers

• If a query fails with a syntax error, fetch the live reserved keyword list: exapump sql "SELECT KEYWORD FROM EXA_SQL_KEYWORDS WHERE RESERVED ORDER BY KEYWORD"

Related Skills

This skill handles core database interaction: connecting, uploading/exporting files, SQL execution, and table design. For UDF development and Script Language Containers, the exasol-udfs skill provides specialized guidance and will activate automatically when relevant.