> ## Documentation Index
> Fetch the complete documentation index at: https://support.configview.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Windmill setup

Configview reads your Windmill instance (the developer automation platform, cloud at `app.windmill.dev` or self-hosted) through its **REST API** (`<your base URL>/api`, described by the OpenAPI spec in the Windmill repository), using an **API token**. Configview sends only `GET` requests. It never creates, changes, runs or deletes anything in Windmill.

You will end up with **2 required secrets** (`WINDMILL_BASE_URL`, `WINDMILL_TOKEN`) and **1 optional secret** (`WINDMILL_WORKSPACES`) in Configview when setup is complete.

> **Scope of this integration today.** Configview answers "who can build and run automations here, and what can they reach": workspace members and their roles, groups, folders and owners, scripts, flows and schedules (owner, language, cron, enabled), the resources that connect to other systems (name and type only), variables (name and secret flag only), the API tokens of the collector's own user, HTTP triggers and how callers authenticate, executions per user, instance users, and a 30-day audit log. It does **not** read script or flow code, resource or variable values, job inputs, outputs or logs, or the arguments of audit events.

***

## What you need

* A Windmill instance and a user who is an admin of the workspaces you want covered
* For instance users and every workspace: a token for an instance superadmin
* The base URL of the instance, with `https`

***

## Step 1: Open the Windmill page in Configview

Open Configview in a second browser tab and leave it open:

`https://{companyname}.configview.com/admin/integrations/windmill`

***

## Step 2: Create the token

1. In Windmill open **Account settings > Tokens** and create a token. Give it an expiry. Windmill has no read-only role, so the token can do whatever its user can do; Configview only ever reads. Use a dedicated service-account user that is an admin (workspace admin for the workspace-level data, superadmin only if you want instance users).
2. In the Configview tab paste it into `WINDMILL_TOKEN` under **Credentials** and save.
3. Save the address of your instance as `WINDMILL_BASE_URL`, for example `https://app.windmill.dev` or `https://windmill.example.com`. Do not add `/api`. Configview refuses plain `http`.
4. Optional: save `WINDMILL_WORKSPACES` as a comma-separated list of workspace ids (for example `prod,staging`) to cover only those. Leave it empty to cover every workspace the token can see.

***

## Step 3: Connect and verify

1. Click **Connect**. Configview creates its tables and schedules every collector at your default run time:

| Script | Notes |
| - | - |
| **Workspaces** | Id, name, owner, domain, dev or fork flag. |
| **Workspace users** | Email, role, admin, operator (run-only), disabled, service account, groups, folders. |
| **Groups** | Members of each group. |
| **Folders** | Owners, creator, last edited. |
| **Scripts** | Path, owner from the path, language, kind, archived, deploy errors. Never the code. |
| **Flows** | Path, owner, last editor, runs on behalf of. Never the steps. |
| **Schedules** | Cron, timezone, enabled, target, run as, last error flag. Never the arguments. |
| **Resources** | Path and type of each stored connection, OAuth state. Never the values. |
| **Variables** | Path, secret flag, expiry. Never the values. |
| **HTTP triggers** | Target, method, authentication method. Never the route path. |
| **Executions by user** | Execution count per user. |
| **API tokens** | The collector user's own tokens: label, scopes, last used, expiry. |
| **Instance users** | Every account on the instance. Needs a superadmin token; skipped otherwise. |
| **Audit log** | Operations in the last 30 days. Needs a workspace admin. |

2. Click **Verify now**. The health check reads both secrets, then `/users/whoami` and the workspace list.

If a check fails:

* **401.** The token is wrong, expired, revoked, or belongs to a different instance.
* **403 or 404.** The token's user cannot read that area. Configview skips that workspace or collector and carries on.

***

## Data Tables

Each table has a `run_at` column and keeps only the newest run.

| Table | Source | Key Columns |
| - | - | - |
| `windmill_workspaces` | `GET /workspaces/list` | workspace\_id, name, owner, domain, parent\_workspace\_id, is\_dev\_workspace |
| `windmill_users` | `GET /w/{workspace}/users/list` | workspace\_id, email, username, user\_role, is\_admin, is\_operator, is\_disabled, is\_service\_account, group\_names, created\_at |
| `windmill_groups` | `GET /w/{workspace}/groups/list` | workspace\_id, group\_name, member\_count, member\_usernames |
| `windmill_folders` | `GET /w/{workspace}/folders/list` | workspace\_id, folder\_name, owners, created\_by, edited\_at |
| `windmill_scripts` | `GET /w/{workspace}/scripts/list` | workspace\_id, script\_path, owner\_kind, owner\_name, language, script\_kind, is\_archived, has\_deploy\_errors, created\_at |
| `windmill_flows` | `GET /w/{workspace}/flows/list` | workspace\_id, flow\_path, owner\_name, edited\_by, edited\_at, on\_behalf\_of\_email, is\_archived |
| `windmill_schedules` | `GET /w/{workspace}/schedules/list` | workspace\_id, schedule\_path, target\_path, cron, timezone, is\_enabled, run\_as\_email, has\_error |
| `windmill_resources` | `GET /w/{workspace}/resources/list` | workspace\_id, resource\_path, resource\_type, created\_by, is\_oauth, is\_expired |
| `windmill_variables` | `GET /w/{workspace}/variables/list` | workspace\_id, variable\_path, is\_secret, is\_oauth, expires\_at, edited\_at |
| `windmill_http_triggers` | `GET /w/{workspace}/http_triggers/list` | workspace\_id, trigger\_path, target\_path, http\_method, authentication\_method |
| `windmill_user_usage` | `GET /w/{workspace}/users/list_usage` | workspace\_id, email, executions |
| `windmill_tokens` | `GET /users/tokens/list` | owner\_email, label, scopes, is\_read\_only, last\_used\_at, expires\_at |
| `windmill_instance_users` | `GET /users/list_as_super_admin` | email, login\_type, is\_super\_admin, is\_devops, is\_disabled |
| `windmill_audit_logs` | `GET /w/{workspace}/audit/list` | workspace\_id, event\_time, username, operation, action\_kind, target\_resource |

***

## Things worth knowing

**Owners come from the path.** Windmill paths start with `u/<user>/` for a personal item or `f/<folder>/` for a folder item. Configview derives the owner from that, because the script list has no creator field.

**Resources show where automations can reach.** A resource is a stored connection (a Postgres database, a Slack workspace, an AWS account). Configview keeps the path and type so you can see which systems are wired in, and whether an OAuth connection has expired. The values, which often hold passwords, are never read even though Windmill's list returns them.

**Tokens are the collector user's own.** Windmill lists only the caller's tokens, so `windmill_tokens` shows the tokens of the account Configview uses, not every token in the workspace.

**HTTP triggers with no authentication.** A trigger whose `authentication_method` is `none` can be started by anyone who knows its URL. Configview lists them.

**Audit log.** The last 30 days, up to 5,000 events per workspace; a warning is printed if the cap is hit. Windmill may limit audit retention on the open-source edition.

**Rate limit.** Windmill documents no numeric API rate limit. Configview paces itself at 120 requests a minute and waits as long as an HTTP 429 asks.

## What isn't collected

* Script and flow code, steps and arguments, schedule arguments
* Resource and variable values, token values
* HTTP trigger route paths and static asset settings
* Job inputs, results and logs (only per-user execution counts)
* The arguments of audit events


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.