# Collect Credentials

> Run the authenticate script to collect credentials from the user via a browser form.

- Skill: `databrickslabs/collect-credentials` (Agent Skill)
- Install (CLI): `npx skillmds@latest add databrickslabs/collect-credentials`
- Raw SKILL.md: https://api.skillmd.com/api/skills/databrickslabs/collect-credentials/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: databrickslabs (https://skillmd.com/u/databrickslabs)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/databrickslabs/collect-credentials

---


# Collect Credentials

## Goal

Run the `authenticate.py` script to collect credentials for the **{{source_name}}** connector. The script starts a local HTTP server that serves a browser form based on the connector spec. The user fills in their credentials in the browser and clicks Save.

## Prerequisites

- `connector_spec.yaml` must already exist at `src/databricks/labs/community_connector/sources/{{source_name}}/connector_spec.yaml`
- Python 3.10+ with project dependencies installed

## Output

- A JSON credentials file at a local path **the user picks**, outside
  the repo. Tests read it via ``CONNECTOR_TEST_CONFIG_PATH=<that path>``
  or by inlining its contents into ``CONNECTOR_TEST_CONFIG_JSON``.

## Steps

1. Ensure the venv and dependencies are ready:
```bash
python3.10 -m venv .venv && source .venv/bin/activate && pip install -e ".[dev]"
```

2. Ask the user where they want the credentials saved (any local
   path; conventionally outside the repo, e.g.
   ``~/secrets/{{source_name}}.json``). Pass that path as ``-o``:
```bash
source .venv/bin/activate && \
  python tools/scripts/authenticate.py -s {{source_name}} -m browser \
    -o ~/secrets/{{source_name}}.json
```

   The ``-o`` flag is required for this skill — never write into the
   repo. Run the script in background (Bash with ``timeout: 0``)
   because it blocks until the user submits the form.

3. Read the background Bash output to extract the URL. The script prints a line like:
```
→ http://localhost:9876
```
The port may differ from 9876 if that port is already in use.

4. Show the URL to the user via `AskUserQuestion`. Ask them to:
   - Open the URL in their browser
   - Fill in their credentials
   - Click Save
   - Confirm here once done

5. Wait for the user to explicitly confirm before proceeding. If the user reports an error, help them debug.

6. After confirmation, verify that the chosen output file exists.
   Tell the user how to point tests at it:
   ```bash
   CONNECTOR_TEST_CONFIG_PATH=<their chosen path> pytest tests/unit/sources/{{source_name}}/
   ```

## Important Notes

- **OAuth connectors**: when `connector_spec.yaml` declares a `connection.oauth` block with an interactive flow (`u2m` / `u2m_per_user`), the script runs the OAuth 2.0 authorization-code flow instead of a plain form — the user provides their `client_id` + `client_secret`, then logs in and authorizes in the browser, and the script obtains the token automatically. The OAuth app in the source must have the script's redirect URI registered. (The script does not run the `m2m` client-credentials grant; for an `m2m` connector the user just supplies `client_id` + `client_secret`.) Otherwise the steps are identical.
- This skill is **interactive** — the script blocks until the user submits the browser form.
- The caller must be able to run Bash in the background and communicate with the user (e.g., via `AskUserQuestion`) while the script is running.
- The script automatically shuts down after the user submits the form.