Skip to content
This repository was archived by the owner on Jun 11, 2026. It is now read-only.
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 9 additions & 6 deletions noom-mcp-server/.env.example
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,10 @@
# Copy this file to .env and fill in the values below.
# DO NOT commit .env to source control.
#
# Authentication: the server opens a browser OAuth flow on first run.
# Authentication: at startup the server checks for a cached Databricks OAuth
# token. If one exists the start is silent; if not, a browser tab opens to
# complete the login flow. On headless machines, run:
# databricks auth login --host https://noom-prod.cloud.databricks.com
# SP credentials are managed by an admin — no config needed here.
# See DEVELOPMENT.md for dev/CI setup.

Expand All @@ -13,8 +16,8 @@ DATABRICKS_HOST=https://noom-prod.cloud.databricks.com
# SQL execution workspace — must match DATABRICKS_HOST
DATABRICKS_MCP_SQL_HOST=https://noom-prod.cloud.databricks.com

# SQL warehouse — all queries are forced to run on this warehouse
# Find it in Databricks UI:
# * SQL Warehouses → Pick your warehouse.
# * In the Overview, you can see the ID by the warehouse name
DATABRICKS_WAREHOUSE_ID=<your-warehouse-id>
# SQL warehouse — all queries are forced to run on this warehouse.
# The default below is the shared Noom production warehouse. Override if
# you need a different warehouse (find its ID in Databricks UI under
# SQL Warehouses → your warehouse → Overview).
DATABRICKS_WAREHOUSE_ID=575c0a43969584a4
21 changes: 8 additions & 13 deletions noom-mcp-server/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -23,34 +23,29 @@ cd databricks-ai-dev-kit/noom-mcp-server
curl -LsSf https://astral.sh/uv/install.sh | sh
```

**3. Authenticate to Databricks with OAuth** (PAT tokens are rejected)

```bash
databricks auth login --host https://noom-prod.cloud.databricks.com
```

**4. Create your `.env` file**
**3. Create your `.env` file**

```bash
cp .env.example .env
```

Open `.env` and set `DATABRICKS_WAREHOUSE_ID` to your SQL warehouse ID.
(Find it in the Databricks UI under **SQL Warehouses → your warehouse → Overview**.)
The `.env` file ships with a default `DATABRICKS_WAREHOUSE_ID`. You can leave it as-is or override it with a different warehouse ID (find it in the Databricks UI under **SQL Warehouses → your warehouse → Overview**).

**5. Install dependencies**
**4. Install dependencies**

```bash
uv sync
```

**6. Verify the server starts**
**5. Verify the server starts**

```bash
uv run --env-file .env python run.py
```

You should see the FastMCP server start without errors. Use `Ctrl-C` to stop it — your MCP client will manage the process from here.
At startup the server checks for a cached Databricks OAuth token. If one exists the start is silent; if not, it opens a browser tab to complete the login flow. On headless machines where a browser can't open, run `databricks auth login --host https://noom-prod.cloud.databricks.com` in a terminal first.

Use `Ctrl-C` to stop it — your MCP client will manage the process from here.

---

Expand Down Expand Up @@ -117,7 +112,7 @@ user's own credentials and are not affected by the patches.

- Python ≥ 3.10
- [uv](https://docs.astral.sh/uv/getting-started/installation/)
- Databricks OAuth session (`databricks auth login` — PAT tokens are rejected at startup)
- Databricks OAuth token (obtained automatically via browser on first start; PAT tokens are rejected)
- READ permission on the `dbrix_mcp_secret` secret scope (provisioned by an admin)

## Environment variables
Expand Down
Loading