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

> ## Agent Instructions
> For agents: install the OpenWork desktop app with `brew install --cask openwork` (macOS) or from https://openworklabs.com/download. Do not run `npx openwork`; the npm package named `openwork` is a different project.
> Connect an agent to OpenWork with the MCP Gateway: `claude mcp add --transport http openwork https://api.openworklabs.com/mcp/agent`. Teams sign up at https://app.openworklabs.com?mode=sign-up.
> Step-by-step agent guide: https://openworklabs.com/docs/start-here/use-openwork-from-an-ai-agent.md
> Connect Slack (any organization admin): https://openworklabs.com/docs/slack/connect-slack.md. OpenWork in Slack (@openwork bot) is private alpha for approved organizations. Setup guide: https://openworklabs.com/docs/slack/overview.md. Agent skill: https://openworklabs.com/.well-known/agent-skills/set-up-openwork-slack/SKILL.md.

# Analytics

> See who on your team uses OpenWork, which skills, plugins and connectors they use, and how your workflow runs went, from Observability › Analytics.

**Observability › Analytics** in OpenWork Cloud shows how your team uses
OpenWork. It has three tabs:

| Tab | What it shows | Plan |
| - | - | - |
| **Usage & adoption** | Active people, sessions, tasks and the models people pick | Enterprise |
| **Plugins & connectors** | How often each plugin, skill and connector is used, by whom, and what fails | Every plan |
| **Workflow Runs** | Your saved workflow and one-off task runs, and whether each one worked | Enterprise |

Analytics only counts things. It never includes prompts, code, file contents,
diffs, secrets or terminal output.

## Who can see it

You need the **View usage analytics** permission. Admins have it by default.
Without it, **Analytics** doesn't appear in the sidebar. See
[Members and roles](https://openworklabs.com/docs/cloud/members-and-rbac.md).

On OpenWork Cloud, **Usage & adoption** and **Workflow Runs** are part of the
Enterprise plan. On other plans, **Usage & adoption** shows an Enterprise notice
instead of numbers, and the **Workflow Runs** tab doesn't appear. Usage is
still recorded, so it shows up as soon as the plan includes it. See
[Pricing and licensing](https://openworklabs.com/docs/start-here/pricing-and-licensing.md).

> [!NOTE]
>
> Numbers appear once members sign in to the OpenWork app with your
> organization and run tasks. Until then, the page shows **No activity yet**.
> See [Sign in from the desktop app](https://openworklabs.com/docs/cloud/get-started.md#sign-in-from-the-desktop-app).

## Usage and adoption

The **Usage & adoption** tab answers "is my team using OpenWork, and does it
work for them?"

* **OpenWork users**: people in the organization, and how many invitations are
  still pending.
* **Active this week**: people with at least one event in the last 7 days, and
  how many were active in the last 30 days.
* **Sessions this week**: sessions in the last 7 days, and in the last 30 days.
* **Tasks this week**: tasks in the last 7 days, and their success rate.

Below the totals, charts show active users, sessions and tasks (completed and
failed) over time. Choose **Daily** for the last 30 days or **Weekly** for the
last 12 weeks. Days are counted in UTC.

The **Models** section lists which models people used in the last 30 days, and
how many sessions used the **Default model** versus a **Manually selected** one.
The last row shows **Avg task duration**, **Tasks completed** and **Tasks
failed** over 30 days.

### Look at one project

Use the project filter (**All projects** by default) to narrow the numbers to
one project, for example to follow a rollout. **OpenWork users** stays the
organization total.

Click **Refresh analytics** to load the latest numbers.

## Plugins and connectors

The **Plugins & connectors** tab shows which parts of your Library people
actually use. It's available on every plan.

1. Choose **Plugins**, **Skills** or **Connectors**.
2. Choose a time range: **7 days**, **30 days** or **90 days**.

At the top you see how many items are **In use**, total **Uses**, **Failed**
uses (plugins and connectors only), and how many are **Not used**. Filter the
list to **All**, **Not used** or **Failing**, or search by name.

What counts as one use:

* **Skills**: each time an agent loads the skill.
* **Plugins**: their skills' loads plus their workflows' runs. **Failed**
  counts failed runs.
* **Connectors**: each tool call made through OpenWork. **Failed** counts
  calls that failed.

Click an item to see **Uses per day**, who uses it and how often, and what
started the uses (**Agents**, **Workflows**, **Automations**, **Apps** or
**Live views**). From there, click **Open in Library** or **Open connector** to
change it. Click a person to see everything they used.

If counting started partway through the time range, the page says **Counting
since** and the date.

Use this tab to find skills nobody uses, and connectors that fail before people
give up on them.

## Workflow runs

The **Workflow Runs** tab lists your own runs of saved workflows and one-off
tasks, newest first. Each run shows its name (or **One-off task**), when it
finished, and **Succeeded** or **Failed**.

Click **Details** on a run to see:

* the workflow's steps, when the run came from a saved workflow
* under **Technical details**: where the run came from (**Source**), the
  **Tool calls** it made, its **Duration**, and the **Error** if it failed

Runs come from OpenWork, [Automations](https://openworklabs.com/docs/start-here/do-work-with-it/workflows.md),
and AI tools connected through the [OpenWork MCP gateway](https://openworklabs.com/docs/start-here/connect-openwork-mcp.md).
The list refreshes every 30 seconds.

## Related

* [Audit logs](https://openworklabs.com/docs/cloud/audit-logs.md): who changed what in your organization.
* [Counting usage](https://openworklabs.com/docs/ai-gateway/counting-usage.md): AI Gateway requests, tokens and
  spend.


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