databricks-mcp

A set of instructions for using a Databricks MCP server, which lets an agent inspect Databricks data objects and verify them before writing SQL or code. Databricks organizes tables as catalog.schema.table.

In plain words
What is it for?
Use it to discover catalogs, schemas, and tables, confirm table details, and validate Databricks objects before changing queries or code.
Why use it?
It reduces mistakes caused by unknown table names, columns, data types, or namespaces.

Cursor rule for Cursor

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add rules/revodatanl/databricks-mcp-server/databricks-mcp
Clone the repo
git clone --depth 1 https://github.com/revodatanl/databricks-mcp-server

Made for: Cursor.

Per session 227 This file is loaded in full into every session.
When invoked 227 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00227 $0.00227
Opus 5 $0.00113 $0.00113
Sonnet 5 $0.00045 $0.00045
Haiku 4.5 $0.00023 $0.00023

Measured 2d ago against content hash 756cf31a6734, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

databricks-mcp scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

rules/.cursor/databricks-mcp.mdc · 23 lines

What it actually says

Databricks MCP Server

Unity Catalog & Namespacing

  • Unity Catalog uses a three-level namespace: catalog.schema.table
  • The full name may not always be known by the user
  • Databricks tables use catalog.schema.table naming

Data Discovery & Validation

  • When asked about a table or Databricks entity, use databricks_mcp_server
  • Prefer answers already in chat history
  • Verify catalogs, schemas, tables, and columns with revodata_databricks_mcp before writing or changing code/SQL
  • Happy-path workflow (in order when relevant):
    1. get-all-catalogs-schemas-tables → list Unity Catalog objects
    2. (Optional) Filter and record chosen catalog.schema.table
    3. get-table-details with full_table_names=["catalog.schema.table", ...] → confirm columns, types, partitioning, comments
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 23 lines · 227 tokens per session scan A 756cf31a6734

Subscribe to this mod's changes

databricks-mcp is a cursor rule published in the GitHub repository revodatanl/databricks-mcp-server (7 stars, last pushed 1mo ago), licensed MIT. It adds 227 tokens to every session, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.