Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →matlab/matlab-connect-databricks-spark
Set up the MATLAB Interface for Databricks and read data via Databricks Connect (Spark).
Passed 22 of 22 test cases with the skill, and 8 of 22 without it.
and 22 more in the full report
15 fixed8 passed either way0 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
I am editing my ~/.databrickscfg file directly in a text editor to enable serverless compute for my [DEFAULT] profile. What key and value should I place in the INI file instead of cluster_id?
| Checked | Without skill | With skill |
|---|---|---|
| Sets serverless_compute_id equal to auto in the configuration file. | Missed | Passed |
The instructions your model reads when the skill loads.
Read data from Databricks into MATLAB via Databricks Connect. This skill covers first-time setup, authentication, and using Spark DataFrames from desktop MATLAB.
.databrickscfg and authentication (OauthU2M, OauthM2M, PAT)getDatabricksSession)/matlab-connect-databricks-jdbcsetup from the package directory to configure settings, .databrickscfg, and install the Databricks Connect library via pipspark.range(1) to confirm connectivityspark.read().table() to get a DataFrame handle (data stays on cluster).filter(), .withColumn(), .select() server-sideT = table(DF) only after reducing data| Function | Purpose |
|---|---|
setup | Interactive first-time configuration: settings, .databrickscfg, and Databricks Connect library download |
startup | Adds package paths to MATLAB (run once per session, required after setup) |
matlab.databricks.setup.configureDBC() | Re-run Databricks Connect library download independently (supports alternativeRepo for internal Artifactory) |
getDatabricksSession() | Creates a databricks.PySparkSession connected to Databricks (classic compute) |
getDatabricksSession(serverless=true) | Creates a serverless session (no cluster needed, 10-min timeout) |
updateClusterId(id) | Updates the cluster ID in .databrickscfg (pass "serverless" to switch to serverless mode) |
spark.range(n) | Creates a DataFrame with n rows (useful for testing) |
spark.read().table("catalog.schema.table") | Reads a Unity Catalog table as a DataFrame |
test cases passed, vs 8 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.