Gemini Cloud Assist Mcp
Connects Claude to Google Cloud Platform through the Gemini Cloud Assist APIs, letting you manage your GCP environment with natural language from the command line. Exposes five core tools: ask_cloud_assist for general GCP questions, design_infra for architecting infrastructure, investigate_issue for troubleshooting, invoke_operation for creating and modifying resources when Agent Actions are enabled, and optimize_costs for spend analysis. Currently in private preview and requires allowlist access from your Google Cloud account team. Note that this has migrated from a local Node.js server to a remote architecture as of v0.8.0, so pin to older versions if you need the legacy implementation during the transition period.
- GitHub stars
- 63
- tools
- 3
- transport
- STDIO
- last updated
- Sep 29, 2025
Tools
3 toolsPublic tool metadata for what this MCP can expose to an agent.
ask_cloud_operatorDelegates multi-step operational work to the Cloud Operator agent: incident and root-cause investigation, rollout review, cost optimization, Terraform best-practices review and generation, and documentation-grounded how-to answers.
Delegates multi-step operational work to the Cloud Operator agent: incident and root-cause investigation, rollout review, cost optimization, Terraform best-practices review and generation, and documentation-grounded how-to answers.
No parameter schema in public metadata yet.
search_operator_sessionsLists your Cloud Operator sessions in a project.
Lists your Cloud Operator sessions in a project.
No parameter schema in public metadata yet.
fetch_operator_sessionFetches a Cloud Operator session by its contextId, including its turns.
Fetches a Cloud Operator session by its contextId, including its turns.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
Gemini Cloud Assist MCP server
[!IMPORTANT] Private Preview Notice The Gemini Cloud Assist MCP server APIs are currently in Private Preview and are behind an allowlist. Please contact your Google Cloud account team to request access.
This repository connects AI coding agents to Cloud Operator, the operator agent of Gemini Cloud Assist. Cloud Operator sees your live Google Cloud resources, logs, metrics, billing and policies, and helps you:
- Troubleshoot incidents and find root causes.
- Review rollouts for readiness and monitor them.
- Design and review infrastructure: Terraform, Kubernetes and Helm, with changes written as reviewable file edits.
- Optimize costs.
- Answer how-to questions, grounded in the Google Cloud documentation.
Details
- Repository
- github.com/googlecloudplatform/gemini-cloud-assist-mcp
- Categories
- AI & LLM Tools, Cloud & Infrastructure
- Package
@google-cloud/gemini-cloud-assist-mcp- Transport
- STDIO
- Registry
- MCP registry, active
More from googlecloudplatform
Related AI & LLM Tools MCP servers
View all
Geminihoutini-ai/gemini-mcpModel Context Protocol server for Google Gemini AI with chat, research, and grounding34stars
Neverinfamous Memory Journal Mcpneverinfamous/memory-journal-mcpA MCP server built for developers enabling Git based project management with project and personal…17stars
Pmll Memory Mcpdrqedwards/ppmPMLL Memory MCP — persistent KV context memory and Q-promise deduplication.14stars
Agent Emailusejunior/email-agent-mcpLocal email connectivity for AI agents — read, draft, send, and organize Outlook mail via MCP7stars
A2dbagentic-eng/a2dbSafe, read-only database access for AI agents. 5 databases, batch queries, SQLGlot.2stars