Focused Tool · Single Binary · Self-Hosted

Self-hosted ETL pipelines.
Define, schedule, transform.

Define data pipelines as config, run them on schedule, transform and move data between sources. The low-code Airbyte alternative for teams who want ETL on their own infrastructure.

Running in under 60 seconds.

Install
curl -fsSL https://stockyard.dev/wrangler/install.sh | sh
Run
DATA_DIR=./data wrangler

  Pipeline runner: http://localhost:9060/api
  Dashboard:       http://localhost:9060/ui
Create a pipeline and run it
curl -s -X POST http://localhost:9060/api/pipelines \
  -H 'Content-Type: application/json' \
  -d '{
    "name": "daily-sync",
    "source": {"type": "http", "url": "https://api.example.com/users"},
    "transform": [{"op": "filter", "field": "active", "eq": true}],
    "destination": {"type": "file", "path": "./output/active-users.json"}
  }'

# Trigger a run manually
curl -s -X POST http://localhost:9060/api/pipelines/daily-sync/run

Three things people use this for.

Sync data between services
Pull records from one API, transform them, push to another. Define it once as config, schedule it, and stop writing one-off scripts that break.
Nightly database ETL
Extract from production read replica, transform for analytics, load into your warehouse. Runs on a cron, logs every run, retries on failure.
CSV/JSON ingestion
Drop files into a directory, Wrangler picks them up, validates the schema, transforms rows, and loads them into your target database or API.

Teams outgrow spreadsheet-based workers tracking around the same time they realize SaaS alternatives want $15 to $50 per seat per month. Wrangler is a single binary that handles the same job without the recurring bill.

Start Wrangler with a port and a data directory. It creates its SQLite database on first run. Create workers through the web interface or POST JSON to the API. Filter and search by keyword.

Define your data pipeline as a series of steps. Wrangler pulls from sources, transforms the data, and pushes to destinations. Runs on a schedule or triggered by webhooks. No managed ETL service required.

Free to run. Pro when you need more.

Free
$0
forever, self-hosted
Best for: solo devs and small projects
  • 2 pipelines
  • Manual trigger only
  • 100 rows per run
  • JSON and CSV sources
  • Console output destination
  • Basic transform (filter, rename)
  • Run history (last 10 runs)
Get started →
Pro
$1.99
per month
Best for: teams and production use
  • Unlimited pipelines
  • Cron scheduling
  • Unlimited rows per run
  • Postgres, MySQL, S3, HTTP sources
  • Any destination (DB, API, file)
  • Custom transforms (JS expressions)
  • Full run history and logs
  • Retry on failure
  • Pipeline versioning

When Wrangler is the right tool.

Wrangler solves one problem: moving data between sources with transforms and scheduling. If you need a simple, self-hosted ETL runner without the overhead of Airbyte or Fivetran, it does the job. Free covers 2 pipelines with manual triggers. Pro ($1.99/mo) adds scheduling, unlimited pipelines, and production-grade sources and destinations.

Part of the Stockyard family.

Stockyard Platform
Wrangler is a standalone tool from the Stockyard family. If you need the full LLM control plane with request tracing, cost control, prompt management, and model routing, or want all 150 tools in one bundle, check out Stockyard Complete or the Stockyard Platform.
Learn about Stockyard →

How Pro gets activated.

When your payment goes through, you'll receive an email at the address you used at checkout. The email contains your license key.

Set it as an environment variable before starting Wrangler:

WRANGLER_LICENSE_KEY=stockyard_your_key_here wrangler

  License:   Pro (your-email@example.com)
  Dashboard: http://localhost:9060/ui

That's it. Pro features unlock immediately on startup. No restart required, no account to log into, no phone-home check. The key is verified locally.

Didn't get the email?
Check your spam folder first. If it's not there after 5 minutes, email hello@stockyard.dev with your payment confirmation and we'll send the key manually.
What counts as a pipeline?
A pipeline is a source → transform → destination config. Each one runs independently on its own schedule.
Does Wrangler require Docker or a database?
No. Single Go binary with embedded SQLite. Run it anywhere you can run a binary.
What sources and destinations are supported?
Free tier: JSON files and CSV. Pro adds Postgres, MySQL, S3, HTTP APIs, and any destination reachable via HTTP or SQL.
How does scheduling work?
Pro tier supports cron expressions. Set schedule: "0 2 * * *" in your pipeline config to run at 2 AM daily.
What's the license?
BSL 1.1. Free to self-host and use. Pro license removes limits and adds scheduling.

Try Wrangler in 30 seconds.

Single binary. Free to start. $1.99/mo for unlimited.

Need the full stack? → Stockyard Complete