SqlXL

Move data between Excel and SQL Server — safely

Stop manually exporting query results, editing spreadsheets, and hoping nothing breaks.

SqlXL lets you bulk insert or update SQL Server data from Excel — with full validation using your existing database constraints.

• No manual scripts
• No fragile copy/paste workflows
• No partial imports or silent data corruption

Built for SQL Server developers and data professionals who rely on Excel but need production-safe results.

---

Why SqlXL exists

If you've ever:

• Exported query results to Excel just to clean them up
• Manually edited data before re-importing it
• Worried about bad data making it into your database
• Written one-off scripts for bulk updates

Then you already know the problem.

SqlXL replaces that entire workflow with something safe, repeatable, and fast.

---

Website: runsqlxl.com

Install

dotnet tool install --global SqlXl

Requires .NET 10.0 or later. Windows only.

First-time setup

sqlxl init --connection "Server=myserver;Database=MyDB;Integrated Security=true;TrustServerCertificate=true;"

Installs SqlXL infrastructure into your database and saves the connection as your default profile. You won't need --connection on subsequent commands.

Want to try it before touching your own database?

sqlxl demo --connection "Server=localhost;Integrated Security=true;TrustServerCertificate=true;"
sqlxl init --connection "Server=localhost;Database=SqlXlDemo;Integrated Security=true;TrustServerCertificate=true;" --profile demo
sqlxl use demo
sqlxl insert --table dbo.Products

Have a spreadsheet but no destination table yet?

sqlxl infer your-data.xlsx --output schema.sql
# review schema.sql, edit if needed, then apply it (sqlcmd, SSMS, etc.)
sqlxl insert --table dbo.YourTable --file your-data.xlsx

(Assumes sqlxl init has been run against your database. infer itself is connectionless — it reads the local xlsx and emits text.)

Commands

sqlxl infer — bootstrap a table from a spreadsheet

Have a spreadsheet but no destination table yet? infer reads the file and emits a CREATE TABLE statement with column types inferred from the data. The DDL is for you to review and run yourself — infer never executes it.

# Print DDL to stdout (pipe-friendly)
sqlxl infer products.xlsx --sheet Products --table Products

# Write DDL to a file plus a JSON inference report
sqlxl infer products.xlsx --sheet Products --table Products --output products.sql --report products.json

After you apply the DDL (any tool — sqlcmd, SSMS, etc.), the same xlsx loads through the validated sqlxl insert pipeline below. No reformatting needed.

sqlxl insert — bulk-insert new rows

# Generate an empty INSERT template
sqlxl insert --table dbo.Products

# Import a filled template
sqlxl insert --table dbo.Products --file Products_insert_20260412.xlsx

No configuration needed. SqlXL inspects the table and scaffolds everything automatically on first use.

sqlxl update — bulk-update existing rows

# Generate a pre-populated UPDATE template (all rows)
sqlxl update --table dbo.Products

# Filter which rows to include
sqlxl update --table dbo.Products --where "CategoryName = 'Electronics'"

# Import a filled template
sqlxl update --table dbo.Products --file Products_update_20260412.xlsx

sqlxl import — custom feature (power users)

For workflows that span multiple tables, or require custom validation logic, configure a BulkOpFeature in your database and reference it by ID:

# Generate template driven by the feature config
sqlxl import --feature 7

# Import a filled template through the configured processing sproc
sqlxl import --feature 7 --file data.xlsx

sqlxl export — export a query to Excel

sqlxl export --query "SELECT * FROM dbo.Products WHERE CategoryID = 3"
sqlxl export --query "SELECT * FROM dbo.Orders" --output orders.xlsx

sqlxl test — smoke-test a table's features

Auto-generates test data and runs it through all configured features for a table. Useful after scaffolding to verify the pipeline end-to-end.

sqlxl test --table dbo.Products
sqlxl test --table dbo.Products --rows 5

sqlxl demo — spin up a demo database

Creates (or resets) a self-contained demo database with sample tables and data. Safe to run repeatedly — drops and recreates.

sqlxl demo --connection "Server=localhost;Integrated Security=true;TrustServerCertificate=true;"

sqlxl llm-context — machine-readable reference for AI agents

Emits a complete, versioned reference document for the installed binary — commands, flags, workflows, gotchas, and the BulkOpFeature schema — so an AI assistant can operate SqlXL fluently without external lookups.

# Text (markdown) — human-readable
sqlxl llm-context

# JSON — structured, schema-validated, agent-friendly
sqlxl llm-context --format json

# Include live DB state: active profile, configured features, domain tables
sqlxl llm-context --format json --include-state

Pass the JSON output to your AI assistant at the start of a session. No DB connection required unless --include-state is used.

How validation works

Data is bulk-copied to a staging table. SQL Server constraints on that table are the validators. If any row fails a constraint, the entire batch is rolled back and row-level error messages are returned to the terminal. No partial imports.

Managing connections

SqlXL stores named connection profiles in ~/.sqlxl/config.json. After sqlxl init, no --connection flag is needed.

# Save additional profiles
sqlxl init --connection "Server=prod;Database=ProdDB;..." --profile prod
sqlxl init --connection "Server=staging;Database=StageDB;..." --profile staging

# Switch the active profile
sqlxl use prod

# List all profiles (shows active with *)
sqlxl connections list

# Remove a profile
sqlxl connections remove staging

Per-command overrides:

# Use a named profile just for this command (doesn't change the active profile)
sqlxl insert --table dbo.Products --profile staging

# Fully explicit connection string (overrides everything)
sqlxl insert --table dbo.Products --connection "Server=...;Database=...;"

Resolution order (highest to lowest priority):

1. --connection flag
2. SQLXL_CONNECTION environment variable
3. --profile flag
4. Active profile in ~/.sqlxl/config.json

Security: Windows Auth connection strings (no password) are stored as plain text — there is nothing to protect. SQL Auth connection strings are automatically encrypted with Windows DPAPI, bound to your user account and machine.

Custom config file location

By default, profiles live in ~/.sqlxl/config.json. You can override that location for any command — useful for CI/CD pipelines, shared team config files, or running multiple SqlXL configurations side-by-side on the same machine.

# Per-command override
sqlxl insert --table dbo.Products --config /shared/team-sqlxl-config.json

# Persistent override via environment variable
export SQLXL_CONFIG=/shared/team-sqlxl-config.json
sqlxl insert --table dbo.Products

Resolution order for the config file location (highest to lowest priority):

1. --config <path> flag
2. SQLXL_CONFIG environment variable
3. ~/.sqlxl/config.json (default)

The override applies to both reads and writes — sqlxl init --config /path/to/file.json will write the new profile to that file.

Requirements

• .NET 10.0 or later
• SQL Server 2019 or later
• Windows (DPAPI credential storage is Windows-only)

License

MIT — see LICENSE

Copyright (c) 2026 Chris Hamilton