Integration · Company data

Connect Claude & ChatGPT to Looker

Ask your AI for last month's revenue by region, which dashboard tracks churn, or to send you a dashboard as a PDF. It can search dashboards, Looks and folders, run saved Looks and queries, build ad-hoc queries from your LookML models and explores, save new Looks, and render dashboards and Looks to files. Data questions get answered without building the query yourself.

Connect Looker free

Free plan, no credit card. Takes about a minute.

Try asking

“Which dashboards do we have about churn?”
The AI searches Looker content for churn and lists the matching dashboards and Looks with their IDs.
“Run the Look 'Weekly signups' and show me the numbers.”
It finds the Look, runs it and returns the data as a table.
“What was revenue by region last month?”
It reads the explore's fields to find the right dimensions and measures, then runs an ad-hoc query with the filters.

What your AI can do

Find content

Search dashboards, Looks and folders, and read a dashboard's tiles and filters or a Look's query.

Run queries

Run saved Looks and queries, or build ad-hoc queries from your models and explores, with results as JSON, CSV, Markdown, XLSX and more.

Save and render

Save new Looks and queries, update Looks, and render dashboards and Looks to PDF, PNG or JPG.

“Save that query as a Look in the Sales folder.”
It saves a new Look with the title, folder and query.
“Render the Executive overview dashboard as a PDF.”
It starts a render task for the dashboard, checks its progress and downloads the finished file.
“Which scheduled deliveries go to [email protected]?”
It searches scheduled deliveries by recipient and lists them.

27 tools for Looker

These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.

create_lookChanges dataSave a new Look. Pass `title`, `folder_id` and either a `query_id` or a `query` (model, view, fields, filters, sorts, limit).
create_queryChanges dataSave a query definition without running it. Pass `model`, `view` (the explore), `fields`, `filters`, `sorts` and `limit`; returns a query ID for run_saved_query or create_look.
get_current_userGet the Looker user the API key belongs to.
get_dashboardGet one dashboard by its ID in `task_id`, with its tiles, filters and layout.
get_exploreGet an explore's fields (dimensions and measures) by `lookml_model_name` and `explore_name`. Use this to find field names before running a query.
get_lookGet one Look by its ID in `task_id`, including its query.
get_modelGet one LookML model by `lookml_model_name`, with its list of explores.
get_queryGet a saved query's definition by its ID in `task_id`.
get_render_resultDownload the finished file of a render task by `render_task_id`.
get_render_taskCheck the status of a render task by its ID in `task_id`.
list_dashboard_filtersList the filters on a dashboard by `dashboard_id`.
list_dashboard_tilesList the tiles (elements) on a dashboard by `dashboard_id`, including each tile's Look or query.
list_folder_dashboardsList the dashboards in a folder by `folder_id`.
list_folder_looksList the Looks in a folder by `folder_id`.
list_modelsList the LookML models you can query, with their explores.
render_dashboardChanges dataStart rendering a dashboard to a file. Pass the dashboard ID in `task_id`, `result_format` (pdf, png, jpg), `width` and `height`, plus optional `dashboard_filters`. Check progress with get_render_task.
render_lookChanges dataStart rendering a Look to an image or PDF. Pass the Look ID in `task_id`, `result_format` (png, jpg, pdf), `width` and `height`. Check progress with get_render_task.
run_lookRun a saved Look by `look_id` and return its data in `result_format` (json, csv, txt, html, md, xlsx, png, jpg). Use `limit` to cap rows.
run_queryChanges dataRun an ad-hoc query and return the data. Pass `model`, `view` (the explore), `fields`, optional `filters`, `sorts`, `limit`, and `result_format` (json, csv, txt, html, md, xlsx).
run_saved_queryRun a saved query by `query_id` and return its data in `result_format` (json, csv, txt, html, md, xlsx).
search_contentSearch Looker content (dashboards and Looks) by `terms`. Narrow with `types` (dashboard, look, lookmldashboard).
search_dashboardsSearch dashboards by `title`, `folder_id`, `user_id` or `description` (use % as a wildcard). Returns dashboard IDs and titles.
search_foldersSearch folders by `name`, `parent_id` or `creator_id`. Returns folder IDs.
search_looksSearch saved Looks by `title`, `folder_id`, `user_id` or `query_id` (use % as a wildcard). Returns Look IDs and titles.
search_schedulesSearch scheduled deliveries by `name`, `look_id`, `dashboard_id`, `user_id` or `recipient`.
search_usersSearch Looker users by `email`, `first_name`, `last_name` or `full_name`.
update_lookChanges dataUpdate a Look by its ID in `task_id`: change its title, description, folder_id or query.

Set up in three steps

1
Pick the app
Create a free PipMCP account and choose Looker from the app list.
2
Paste your key
Paste a Looker Client ID and Client secret, and enter your Looker host name without https:// (for example acme.cloud.looker.com). An admin enables API keys for your user under Admin → Users (edit the user → API keys → New API key). After that you can also manage them from your Account page.
3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
1. Click your name, then Settings › Connectors › Add custom connector.
2. Paste your link as the Remote MCP server URL.
3. Switch it on from the + menu in a chat.

Questions

What can the AI do in Looker?

It can search and read dashboards, dashboard tiles and filters, Looks, folders, LookML models, explores, saved queries, scheduled deliveries and users. It can run Looks and queries, build ad-hoc queries, save new Looks and queries, update Looks, and render dashboards and Looks to files.

Does the AI see my Looker credentials?

No. Your Looker credentials are encrypted at rest and never shown to the AI. After you save them, they are not shown again, not even to you. The AI only sees the results of the tools it calls.

Can I control what the AI is allowed to do?

Yes. You choose which tools are switched on, so you can start with searching and running existing Looks. Saving and updating Looks changes content your colleagues see, so these can require your approval before they run. Every tool call is logged.

Does it work with ChatGPT?

Yes. In ChatGPT, open Settings › Apps & Connectors › Advanced, turn on Developer mode and add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.

What does it cost?

PipMCP has a free plan with no credit card required. Paid plans bill per completed task. You also need your own Looker account.

What data can the AI see in Looker?

The AI works through the Looker user the API key belongs to. It can only see the models, explores, folders and content that this user has access to in Looker. To limit what it can reach, create the API key on a user with narrower permissions.