Google Cloud

Looker

Included as a governed BI and semantic-modeling surface whose managed MCP can query LookML-backed business data and, when administrators enable the relevant tools, create personal Looks or dashboards and update LookML project files under the signed-in user's permissions.

Analytics, SEO & experimentationSystem of intelligenceAgent-accessible SaaS
An illustration of access surfaces surrounding a governed system.
System roleSystem of intelligence
Access maturityAgent-accessible SaaS
MCP supportOfficial MCP server
Reference updatedJul 10, 2026
BYO-UI reviewedJul 22, 2026

Concrete capability record

Data, retrieval, actions, identity, and operating limits

A field-by-field summary of what the reviewed first-party references actually support. Publisher update dates and BYO-UI review dates are shown separately below.

01 · Records and state owned

LookML projects and files, models and Explores, inline queries, generated SQL and Explore URLs, Looks, dashboards, personal folders, and query parameters exposed by the enabled managed-MCP tools.

02 · Data you can retrieve

Search projects, project files, Looks, and dashboards; inspect LookML and query parameters; build semantic-model queries or SQL; generate Explore URLs; and run Looks or dashboards when those tools are enabled.

03 · Actions and write paths

Enabled managed-MCP tools can create a Look or dashboard in the authenticated user's personal folder and update a LookML project file. Broader Looker API mutations are not automatically MCP tools.

04 · Authentication and permissions

The managed endpoint uses OAuth 2.1 and standard Looker sign-in. During Preview an administrator manually registers the client's OAuth credentials; calls inherit the authenticated user's Looker roles, model access, permissions, and folder access, while an Admin separately controls which MCP tools are enabled.

05 · Monitoring, approval, and recovery

All managed-MCP tools are disabled by default. A Looker Admin can enable tools individually or enable all, which can automatically expose tools added later. Looker applies user permissions, VPC Service Controls and CMEK controls, and records actions in System Activity; Google Cloud core instances also use Cloud Audit Logs.

06 · Limits and caveats

Managed MCP is Preview and supports Looker-hosted Google Cloud core and original instances, not customer-hosted deployments. It has no fine-grained OAuth scopes or dynamic client registration, fixed preview capacity and timeouts, standard API quota consumption, delayed tool-setting propagation, and an IP-allowlist incompatibility on Looker (original).

Agent access

MCP support

Official MCP serverLooker-managed remote MCPPreview

Support: Official MCP server

Read scope: Invoke only administrator-enabled tools to search Looker content and LookML, construct semantic queries or SQL, generate Explore links, and execute Looks or dashboards within the user's existing access.

Write scope: Managed-MCP writes currently documented are personal-folder Look or dashboard creation and LookML project-file updates; each requires its tool to be enabled and the user to possess the underlying Looker permission.

Authentication: OAuth 2.1 at the instance `/mcp` endpoint, with standard Looker user sign-in and administrator registration of the client's OAuth credentials during Preview.

Approval boundary: Before `make_look`, `make_dashboard`, or `update_project_file`, show the target folder or project file, proposed content, and authenticated user; require confirmation for project-file changes. Treat the Admin `Enable all` option as approval for present and future tools, not only today's catalog.

Protocol does not erase product boundaries.

Confirm the current tool catalog, plan, region, scopes, rate limits, terms, and write behavior before implementation.

Editorial assessment

Access-maturity dimensions

A comparative architecture lens—not a quality score, market ranking, or buying recommendation. Scale: 1–5.

Data access4
Action access4
Event access1
Identity4
Governance4
Agent access3
UI extensibility3
Portability4
Observability3
Documentation4

Proposed customer-shaped experiences

What customers could create on top.

The output could be an export, report, graph, artifact, application, agent, workflow, or downstream feed. These proposals are derived from documented access—not claims that Google Cloud ships them.

Single-platform patterns

  • Governed semantic-metrics explorer
  • Executive insight brief
  • Experiment review board
  • Anomaly investigation room
Use in the brief builder →

Multi-platform compositions

  • Looker + warehouse or semantic layer: executive insight brief
  • Looker + CRM or lifecycle platform + work-management system: cross-system decision workspace
Explore composition recipes →

Evidence and dates

First-party references, with publisher and review dates separated

“Publisher updated” is shown only when the page exposes an update date. “BYO-UI reviewed” records when this research checked the reference. A missing publisher date is reported as missing—not replaced with the review date.

5 recorded sources
MCPRepresentative source
https://docs.cloud.google.com/looker/docs/mcp ↗

Official Google Cloud docs confirm built-in managed MCP in preview and admin tool allowlisting.

Publisher updated
Jul 10, 2026
BYO-UI reviewed
Jul 12, 2026

Targeted first-party MCP/agent-role review; broader API inventory retained

Developer docsSource inventory
https://cloud.google.com/looker/docs ↗

First-party developer documentation or platform overview.

Publisher updated
Jul 10, 2026
BYO-UI reviewed
Jul 10, 2026

Inventory source: structurally normalized; content was not individually reopened in this pass.

HomepageSource inventory
https://cloud.google.com/looker ↗

First-party product homepage used to confirm product identity and current positioning.

Publisher updated
Not stated by publisher
BYO-UI reviewed
Jul 10, 2026

Inventory source: structurally normalized; content was not individually reopened in this pass.

Looker query and SQL RunnerTargeted review
Looker API Query methods ↗

Looker APIs can create and run governed model queries and SQL Runner queries. Effective access depends on Looker models, connection credentials, user permissions, and SQL Runner permission.

Publisher updated
Jun 9, 2026
BYO-UI reviewed
Jul 12, 2026

Targeted first-party access review

Verified fact

Tied to cited first-party evidence reviewed for this profile.

Source inventory

Official links recorded for deeper research but not necessarily reopened endpoint by endpoint.

Editorial assessment

System role, maturity interpretation, and architectural boundary.

Proposed design

Interface patterns and compositions—not vendor product claims.