DEV Community: curatedmcp

Perspective AI: Replace Forms with Conversational Data Collection in Your AI Workflows

curatedmcp — Sat, 01 Aug 2026 10:26:11 +0000

Perspective AI: Replace Forms with Conversational Data Collection in Your AI Workflows

Static forms are friction. They drop conversion rates, lose context, and force users into rigid data structures that don't match how they actually think. Perspective AI's MCP server lets you embed an adaptive AI concierge directly into your Claude, Cursor, or Windsurf workflows—turning form-filling into natural conversation while automatically structuring the data that matters.

What It Does

Perspective AI MCP enables your AI agent to conduct intelligent, context-aware interviews instead of deploying dumb forms. The server handles the conversational logic—asking clarifying questions, understanding nuance, and adapting follow-ups based on previous answers—then outputs clean, structured data and triggers downstream automation.

For developers, this means:

Automatic qualification and routing without predefined decision trees
Rich context capture that preserves the "why" alongside the "what"
Reduced drop-off by replacing multi-step forms with natural dialogue
Structured outputs ready for your backend systems, CRMs, or decision engines

Whether you're building customer intake flows, research surveys, user interviews, or evaluation systems, Perspective AI handles the messy middle—the part where humans need to ask follow-up questions and think contextually.

How to Install

npx mcp-remote https://getperspective.ai/mcp

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "perspective-ai": {
      "command": "npx mcp-remote https://getperspective.ai/mcp"
    }
  }
}

Then reference it in your Claude conversations or integrate it into Cursor or Windsurf workflows. Full setup guides available in the docs.

Real-World Use Cases

Customer qualification interviews — Replace a 5-field signup form with an AI concierge that asks contextual questions, understands the customer's actual problem, and routes them to the right team or product tier.
User research at scale — Conduct adaptive surveys where follow-up questions branch based on answers, capturing richer insights than checkbox forms while reducing fatigue.
Incident triage & support escalation — Have Claude automatically interview users about their issue, dig deeper where it matters, and surface structured severity/context data to your support team without manual forms.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

DataForSEO MCP Server: Connect AI Agents to Enterprise SEO Data in Real Time

curatedmcp — Fri, 31 Jul 2026 10:26:11 +0000

DataForSEO MCP Server: Connect AI Agents to Enterprise SEO Data in Real Time

DataForSEO MCP Server bridges the gap between AI assistants and comprehensive SEO intelligence. This Model Context Protocol implementation opens DataForSEO's nine APIs to Claude, Cursor, Windsurf, and other MCP-compatible agents—giving your AI workflows direct access to real-time SERP data, keyword research, backlink analysis, and on-page crawl metrics without manual API calls.

What It Does

The server exposes DataForSEO's full API suite through a standardized MCP interface. Instead of writing HTTP requests, your AI agent can natively query search engine results, discover high-intent keywords, analyze competitor backlinks, crawl websites for SEO issues, and pull business intelligence—all in natural conversation.

Key capabilities include:

SERP API: Live Google, Bing, and Yahoo search results with rankings and CTR estimates
KEYWORDS_DATA API: Search volume, CPC, trends, and clickstream patterns
ONPAGE API: Automated website crawling with performance metrics
BACKLINKS API: Referring domain analysis, anchor text distribution, link quality scoring
DOMAIN ANALYTICS API: Traffic estimates, tech stack detection, Whois data
CONTENT ANALYSIS API: Brand monitoring, sentiment tracking, citation detection
AI_OPTIMIZATION API: Conversational keyword insights and real-time LLM benchmarking

This eliminates context switching—your agent stays focused on analysis while delegating data collection to a trusted source.

How to Install

npx dataforseo-mcp-server

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "dataforseo-mcp-server": {
      "command": "npx dataforseo-mcp-server",
      "env": {
        "DATAFORSEO_LOGIN": "your-email@example.com",
        "DATAFORSEO_PASSWORD": "your-api-password"
      }
    }
  }
}

Restart Claude Desktop. The server is free to install; DataForSEO API pricing depends on usage.

Real-World Use Cases

Competitive keyword analysis: Ask Claude to fetch top-ranking keywords for a competitor domain, analyze search volume vs. difficulty, and recommend content opportunities—all in one prompt.
SEO audit automation: Have your agent crawl a website, identify on-page issues (missing meta tags, crawl errors, performance gaps), and generate a prioritized fix list.
Content strategy research: Pull trending keywords via AI_OPTIMIZATION API, cross-reference backlink profiles of top competitors, then generate a data-backed content calendar.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

HubSpot MCP: Connect Your AI Agent Directly to HubSpot's CRM, Marketing, and Service Tools

curatedmcp — Thu, 30 Jul 2026 10:26:10 +0000

HubSpot MCP: Connect Your AI Agent Directly to HubSpot's CRM, Marketing, and Service Tools

The HubSpot MCP server is the official bridge between AI agents (Claude, Cursor, Windsurf) and HubSpot's CRM ecosystem. Instead of manually switching between your AI tool and HubSpot's UI, your agent can now read and write directly to contacts, deals, tickets, and campaigns. It's built for teams that want automation without leaving their development environment or AI chat.

What It Does

HubSpot MCP unlocks a set of capabilities that turn your AI agent into a CRM operator:

Contact & company management — create, update, and query customer records programmatically
Deal and pipeline automation — move deals through stages, update revenue forecasts, track sales metrics
Support ticket handling — create tickets, manage workflows, escalate issues
Marketing analytics — query email performance, campaign metrics, and list engagement
Sales intelligence — access pipeline reports, revenue data, and task generation

Think of it as giving Claude or your Cursor agent read-write access to your entire customer data layer. You can ask your agent to "pull all deals closing this quarter and generate a summary," or "create a support ticket from this customer issue and assign it to the right team."

How to Install

Install via npm:

npx -y @hubspot/mcp

Then configure it in your Claude Desktop config file:

{
  "mcpServers": {
    "hubspot-mcp": {
      "command": "npx -y @hubspot/mcp",
      "env": {
        "HUBSPOT_API_KEY": "your-hubspot-private-app-token"
      }
    }
  }
}

You'll need a HubSpot private app token from your account settings. Full setup docs here.

Real-World Use Cases

Sales workflow automation — Ask Claude to "find all deals stuck in negotiation for >30 days, then draft follow-up emails and create reminder tasks for each owner"
Customer support triage — Have your agent auto-create tickets from Slack messages, categorize them by urgency, and route them to the right queue
Marketing reporting — Generate weekly email performance summaries by querying campaign metrics, then post results to a Slack channel via a separate integration

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Databricks MCP: Give Claude Direct Access to Your Lakehouse

curatedmcp — Wed, 29 Jul 2026 10:26:11 +0000

Databricks MCP: Give Claude Direct Access to Your Lakehouse

The official Databricks MCP server bridges the gap between AI agents and your data lakehouse. Instead of describing your schema or manually copying query results, Claude can now execute notebooks, run SQL directly against Delta Lake, manage clusters, and inspect data lineage—all conversationally.

This is particularly useful if you're building AI workflows that need to touch production analytics or data engineering assets. Your agent can query tables, check job status, and pull insights without leaving the conversation.

What It Does

Databricks MCP unlocks several concrete capabilities for AI agents:

Execute notebooks and fetch results — tell Claude to run a specific notebook, and it returns the output directly
Query Delta Lake and Unity Catalog — ask questions about your lakehouse data; Claude runs the SQL
Manage compute — inspect cluster status, scale resources, or review job history
Inspect schemas and lineage — Claude can explore table structure and data provenance
Access MLflow — query model registry, review experiment runs, and log metrics programmatically
Generate analytics — run SQL, build insights, and create simple visualizations

The server handles authentication via your Databricks workspace token, so no credential juggling required once configured.

How to Install

Install the server via npm:

npx -y @databricks/mcp

Add it to your Claude Desktop config:

{
  "mcpServers": {
    "databricks-mcp": {
      "command": "npx -y @databricks/mcp",
      "env": {
        "DATABRICKS_HOST": "https://your-workspace.cloud.databricks.com",
        "DATABRICKS_TOKEN": "your-personal-access-token"
      }
    }
  }
}

See the official docs for detailed setup across all platforms.

Real-World Use Cases

Automated data quality checks: Ask Claude to run your validation notebook nightly, then alert you if row counts or schema mismatches appear
Self-service analytics: "Query revenue by region from the silver tables and explain trends" — Claude executes the SQL and interprets results
ML experiment tracking: Have Claude log hyperparameters and metrics to MLflow during training, or query past experiments to compare models

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

PagerDuty MCP: Let AI Help Your Team Respond to Incidents Faster

curatedmcp — Tue, 28 Jul 2026 10:26:11 +0000

PagerDuty MCP: Let AI Help Your Team Respond to Incidents Faster

Incident response doesn't need to be fully manual anymore. PagerDuty MCP is the official integration that gives Claude, Cursor, and other AI agents direct access to your incident workflows—letting them create incidents, update on-call schedules, manage alerts, and coordinate response actions without switching tools.

What It Does

PagerDuty MCP bridges your incident management system and AI agents. Instead of asking your team to manually log incidents or chase down who's on-call, you can now let Claude or Cursor handle that directly.

The server unlocks these capabilities:

Create and update incidents with custom urgency, priority, and assignment
Query on-call schedules to find the right person for the job
Manage escalation policies and teams programmatically
Review alert history and response timelines to understand incident patterns
Annotate incidents with notes and context as response unfolds
Trigger incident actions from natural language commands in your AI conversations

This is built for ops teams who want AI to accelerate their incident response cycle—reducing the time between detection and action, and keeping responders aligned without manual overhead.

How to Install

Install via npm:

npx -y @pagerduty/mcp

Then add it to your Claude Desktop claude_desktop_config.json:

{
  "mcpServers": {
    "pagerduty-mcp": {
      "command": "npx -y @pagerduty/mcp",
      "env": {
        "PAGERDUTY_API_KEY": "your-api-key-here"
      }
    }
  }
}

Get your API key from PagerDuty's developer portal.

Real-World Use Cases

Auto-escalate critical alerts: Ask Claude to create a high-urgency incident and page the on-call engineer in one command when your monitoring system detects a production outage.
Build runbook workflows: Use Cursor to generate incident response steps, then have Claude automatically update the incident with resolution details and responder notes.
Query incident trends: Let Claude analyze your alert history to identify patterns—which services fail most often, which teams respond fastest—then generate a summary for your next retro.

Free to use. Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

MCP Ecosystem Week 31: When Official Integrations Become Your Biggest Allowlist Decision

curatedmcp — Mon, 27 Jul 2026 10:26:21 +0000

MCP Ecosystem Week 31: When Official Integrations Become Your Biggest Allowlist Decision

The MCP ecosystem continues to consolidate around official integrations from major vendors — and that consolidation is exactly where governance matters most. This week, zero new servers were reviewed, but the usage data tells a clearer story: your developers are converging on a small set of high-trust integrations. That's good for security surface area. It's bad if you haven't formally decided whether to allow them.

This Week in MCP

No new servers entered the catalog this week, but that's not stagnation — it's signal that platform teams are focusing on vetting and deploying the existing 74 risk-classified servers rather than chasing novelty. If you're running CuratedMCP, this is the moment to audit which of those 74 are actually in use across your developer fleet, versus which are sitting in your policy library untouched.

The real governance work happens now: moving from "we have a catalog" to "we have a decision for every server our developers want."

On the Radar

The five most-viewed servers this week are all official integrations from major platforms, and they warrant explicit allowlist decisions:

GitHub Copilot MCP (98K views) — Direct integration with GitHub Copilot's code intelligence. Before allowlisting: confirm your SSO grants the right GitHub org scope, and audit whether Copilot telemetry is routed through your compliance boundary.

OpenAI MCP (87K views) — Access to GPT-4o, DALL-E, Whisper, and Embeddings. This is a supply-chain node: it requires OpenAI API keys. Enforce key rotation policy and segregate API spend across teams if you're tracking it.

Figma MCP (82K views) — Design file and token access in your AI workflow. Data classification concern: confirm whether design files are appropriately scoped in Figma's RBAC, and whether your agents should have access to all components or a subset.

GitHub MCP (76K views) — Repo, issue, PR, and workflow management. This is a high-impact surface: agents writing to repos need audit trails. Enforce branch protection and require pull request review even for AI-generated code.

Anthropic Claude MCP (76K views) — Nested Claude reasoning. Governance question: are you comfortable with intra-Claude API calls, and do you want to meter them separately from first-party Claude usage?

Governance Take

Here's the real risk this week: allowlist fragmentation across your IDE and agent fleet.

You've likely deployed Claude Code, Cursor, Windsurf, and GitHub Copilot across your team. Each has its own MCP integration pathway. Each developer can independently add servers. And unless you're actively enforcing policy at the machine level — not just documenting it — you have no view into which servers are actually running in which client.

The five most-viewed servers aren't controversial individually. But "GitHub Copilot MCP is allowed in Cursor" and "GitHub Copilot MCP is blocked in Claude Code" is a policy debt that compounds weekly. Platform teams we work with find, on audit, that their allowlist has drifted across clients within months.

Second: token spend and audit-log visibility aren't the same thing. With TokenShield, you get a live ledger of Claude spend broken down by machine and user. That visibility is the foundation for governance — you can see where your spend is concentrating, and whether it maps to your allowlist decisions. If 60% of Claude tokens are flowing through a server you haven't reviewed, that's a policy gap. TokenShield surfaces that gap in days, not months.

This week: run an audit across your fleet. Log into CuratedMCP and compare your active allowlist against actual usage. The gap is where governance happens.

Govern MCP usage across your team with CuratedMCP — or scan your own stack free at https://www.curatedmcp.com/auditor.

Splunk MCP: Let Your AI Agent Query Observability Data and Triage Incidents

curatedmcp — Mon, 27 Jul 2026 10:26:11 +0000

Splunk MCP: Let Your AI Agent Query Observability Data and Triage Incidents

Splunk MCP is the official Model Context Protocol server that connects Claude, Cursor, and other AI agents directly to your Splunk instance. Instead of manually digging through dashboards and log searches, your AI can now query observability data, pull alert timelines, correlate security events, and surface root causes from massive datasets—all in natural conversation.

The server acts as a bridge between AI agents and Splunk's data layer. You can ask Claude to search logs across services, inspect metrics and traces, retrieve alert histories, and cross-reference operational events with security incidents. For SRE and SecOps teams, this means faster incident triage and better context gathering without context-switching to the Splunk UI.

What It Does

Splunk MCP unlocks these core capabilities:

Run Splunk queries and retrieve event results directly from your search head
Inspect logs, metrics, and traces across distributed services
Access alert and incident timelines to understand event history
Search dashboards and saved searches without manual lookups
Manage Splunk apps and data inputs programmatically
Correlate security and operational events to find patterns humans might miss

This is especially powerful for incident response. Instead of copying log snippets into chat, Claude can pull relevant events, correlate them with alert data, and suggest root causes—all while staying within the context window of your conversation.

How to Install

Install via npm:

npx -y @splunk/mcp

Add to your Claude Desktop config (claude_desktop_config.json):

{
  "mcpServers": {
    "splunk-mcp": {
      "command": "npx -y @splunk/mcp",
      "env": {
        "SPLUNK_URL": "https://your-splunk-instance:8089",
        "SPLUNK_AUTH_TOKEN": "your-bearer-token"
      }
    }
  }
}

For Cursor and Windsurf setup guides, check the full documentation at CuratedMCP.

Real-World Use Cases

Incident triage in real time: Ask Claude to search for error spikes in your logs, then have it correlate those errors with triggered alerts and recent deployments to identify the root cause.
Cross-service debugging: When a user reports slow requests, have your AI query traces across multiple services, pull matching metrics, and surface which service introduced the latency.
Security event correlation: Use Claude to search for failed login attempts, then correlate them with network traffic anomalies and firewall blocks to identify attack patterns.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

ServiceNow MCP: Automate ITSM Workflows Directly From Claude

curatedmcp — Sun, 26 Jul 2026 10:26:11 +0000

ServiceNow MCP: Automate ITSM Workflows Directly From Claude

ServiceNow MCP is the official integration that connects AI agents to ServiceNow's IT service management platform. If you're running operations and want Claude or Cursor to handle incident creation, change management, and service catalog queries without leaving your IDE or chat interface, this is the tool.

What It Does

ServiceNow MCP exposes your ServiceNow instance to AI agents as a set of callable tools. Instead of switching between ServiceNow's web interface and your development environment, you can describe what needs to happen in natural language—and the MCP server translates that into ServiceNow API calls.

Core capabilities include:

Incident & Problem Management: Create, update, and search incidents. Automatically route tickets based on rules you define.
Change Requests: Propose and track changes with full workflow integration.
Knowledge & Service Catalog: Query KB articles and catalog items to surface solutions or provision services.
Approvals & SLAs: Manage workflow states, track SLAs, and automate approval chains.
Asset & Configuration Data: Query CMDB for context on infrastructure and dependencies.
Analytics: Access service desk metrics and trend data for reporting.

This is built for ops teams that want to reduce manual ticket work and let AI handle the repetitive parts of service lifecycle management.

How to Install

npx -y @servicenow/mcp

Add to your Claude Desktop config:

{
  "mcpServers": {
    "servicenow-mcp": {
      "command": "npx -y @servicenow/mcp",
      "env": {
        "SERVICENOW_INSTANCE": "your-instance.service-now.com",
        "SERVICENOW_API_KEY": "your-api-key"
      }
    }
  }
}

Grab your API credentials from ServiceNow's developer portal and you're ready.

Real-World Use Cases

Bulk Incident Creation from Alerts: Feed Claude a list of production alerts and have it auto-create incidents in ServiceNow with proper assignment, priority, and tags—no manual ticket entry.
Change Request Drafting: Describe a deployment in Cursor, and let the MCP server scaffold a full change request with description, risk assessment, and approval workflow.
Knowledge Base Search During Incident Response: When troubleshooting, ask Claude to search your ServiceNow KB in real-time and surface relevant articles without breaking context.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Okta MCP: Give Claude Safe Access to Your Identity Infrastructure

curatedmcp — Sat, 25 Jul 2026 10:06:32 +0000

Okta MCP: Give Claude Safe Access to Your Identity Infrastructure

Identity management is tedious—creating users, rotating app access, investigating suspicious logins. Okta MCP lets you hand these tasks to Claude while keeping your authentication layer locked down.

This official Okta server bridges your identity platform to AI agents. Instead of manual API calls or UI clicking, you can ask Claude to provision users, audit group memberships, reset passwords, or analyze authentication events. It's built for security and IT teams who need faster incident response without sacrificing control.

What It Does

Okta MCP unlocks six core capabilities:

User & group management: Create, update, deactivate users; manage group assignments and roles
App access: Assign applications, configure OAuth/OIDC clients, query sign-on rules
Authentication insights: Inspect MFA status, review auth events, detect anomalies
Password & recovery: Automate password resets and recovery workflows
Policy enforcement: Query and apply access policies programmatically
Security response: Empower AI to react to identity incidents in real time

All requests stay within your Okta tenant. Claude gets no direct API key—just scoped, auditable actions.

How to Install

npx -y @okta/mcp

Add to your Claude Desktop config:

{
  "mcpServers": {
    "okta-mcp": {
      "command": "npx -y @okta/mcp",
      "env": {
        "OKTA_DOMAIN": "https://your-domain.okta.com",
        "OKTA_API_TOKEN": "your_api_token_here"
      }
    }
  }
}

Replace OKTA_DOMAIN and OKTA_API_TOKEN with your actual values. See full setup docs at developer.okta.com/docs/mcp.

Real-World Use Cases

Onboarding at scale: Ask Claude to create a new employee, assign them to the correct groups, grant app access, and send a summary—all in one prompt.
Incident response: "Check the last 50 auth events for failed MFA attempts, flag suspicious IPs, and reset passwords for affected accounts."
Policy audits: "List all users with admin roles and compare their app assignments against our security policy."

Free to use. CuratedMCP has full install guides for Claude Desktop, Cursor, Windsurf, and more.

IBM Cloud MCP: Control IBM Cloud infrastructure and Watson AI directly from Claude

curatedmcp — Fri, 24 Jul 2026 10:06:33 +0000

IBM Cloud MCP: Control IBM Cloud infrastructure and Watson AI directly from Claude

Managing IBM Cloud resources usually means switching between dashboards, CLIs, and consoles. IBM Cloud MCP flips that—it brings your entire IBM Cloud environment into Claude, Cursor, or Windsurf as native tools your AI agent can call directly.

This is the official IBM integration. You get programmatic access to compute (Kubernetes, VMs, serverless), Watson AI services, databases (Cloudant, Db2), identity policies, and monitoring. No API wrapping or third-party approximations. If IBM Cloud supports it, your agent can do it.

What It Does

IBM Cloud MCP exposes your cloud resources as callable functions. Ask Claude to "provision a Kubernetes cluster in Dallas," "query my Cloudant database," or "check security group rules"—and it happens. The server handles authentication, translates your requests to IBM Cloud APIs, and streams back results.

For enterprise teams, this unlocks real workflow acceleration: infrastructure-as-chat for DevOps, automated compliance audits, intelligent database queries, and Watson AI model deployment without leaving your editor. Your agent can inspect logs, manage access policies, and scale applications across regions—all conversationally.

It's built for hybrid cloud scenarios where your workloads span on-premises and IBM Cloud, and where security and access control matter.

How to Install

npx -y @ibm/mcp

Add to your Claude Desktop config file:

{
  "mcpServers": {
    "ibm-cloud-mcp": {
      "command": "npx -y @ibm/mcp",
      "env": {
        "IBM_CLOUD_API_KEY": "your-api-key-here"
      }
    }
  }
}

Grab your API key from IBM Cloud Identity & Access, restart Claude, and you're live.

Real-World Use Cases

Infrastructure audit in seconds: Ask Claude to list all running Kubernetes clusters, their node counts, and security policies—then suggest cost optimizations.
Database troubleshooting: Query Cloudant or Db2 directly from your chat, inspect slow queries, and have Claude suggest index improvements based on actual schema.
Compliance checks: Automate regular scans of IAM policies, security groups, and encryption settings across your IBM Cloud account—flag violations before audit time.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Adobe MCP: Connect Claude to Photoshop, Acrobat, and Experience Cloud

curatedmcp — Thu, 23 Jul 2026 10:06:34 +0000

Adobe MCP: Connect Claude to Photoshop, Acrobat, and Experience Cloud

Adobe just released an official MCP server that lets AI agents control your creative workflows programmatically. If you're building tools for design or marketing teams, this changes how you can automate asset creation, PDF workflows, and campaign analytics—without leaving your Claude or Cursor chat.

What It Does

Adobe MCP exposes core Adobe services through a standardized protocol. Your AI agent can now:

Create and edit images in Photoshop—generate designs from prompts, apply effects, layer text, and export at scale
Manage PDFs in Acrobat—generate documents, merge files, extract data, apply digital signatures
Query marketing data from Experience Cloud—pull campaign performance, audience insights, and analytics
Access Creative Cloud assets—retrieve libraries, apply templates and brand styles, manage version control
Export deliverables with live URLs—no manual download steps

This is the bridge between generative AI and Adobe's ecosystem. Instead of Claude describing what to build, it can now build it, then hand you a ready-to-use URL.

How to Install

First, install the server via npm:

npx -y @adobe/mcp

Then add it to your Claude Desktop config (claude_desktop_config.json on macOS/Linux, or the Windows equivalent):

{
  "mcpServers": {
    "adobe-mcp": {
      "command": "npx -y @adobe/mcp"
    }
  }
}

Restart Claude Desktop. You'll now see Adobe tools available in your MCP menu.

Real-World Use Cases

Batch social media content: Prompt Claude to generate 10 Instagram post designs with your brand colors and fonts, then export all as web-ready files in one pass
PDF automation: Build a bot that takes user submissions, generates a branded PDF report with charts from Experience Cloud, and emails it—all triggered from a webhook
Dynamic templates: Use Claude to fill Adobe template libraries with campaign-specific copy and imagery, then export for email or web without touching Photoshop

Adobe MCP is free and officially supported. Check the developer docs for the full API reference.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.

Oracle Cloud MCP: Automate OCI Infrastructure from Your AI Agent

curatedmcp — Wed, 22 Jul 2026 10:55:22 +0000

Oracle Cloud MCP: Automate OCI Infrastructure from Your AI Agent

Enterprise teams running Oracle Cloud Infrastructure (OCI) now have a direct path to AI-assisted cloud operations. Oracle Cloud MCP is the official Model Context Protocol server that connects AI agents like Claude and Cursor to your OCI environment—letting you provision compute, manage databases, handle storage, and configure security entirely through natural language.

What It Does

This is a native integration between your AI agent and OCI's full service catalog. Instead of switching between your editor, the OCI console, and CLI tools, you describe what you need and the server handles the REST API calls.

The server unlocks these capabilities:

Compute: Provision and manage VM instances with specific shapes and configurations
Storage: Create and manage Object Storage buckets, upload/download objects
Databases: Query Autonomous Database instances and traditional DB systems
Networking: Inspect Virtual Cloud Networks (VCNs), security lists, and firewall rules
Identity & Access: Create users, manage IAM policies, enforce least-privilege access
Audit & Security: Pull audit logs and security reports for compliance tracking

For teams managing multi-instance deployments or complex database configurations, this eliminates repetitive console navigation and CLI command lookups.

How to Install

Install via npm:

npx -y @oracle/mcp

Then add it to your Claude Desktop configuration (find this file at ~/.config/Claude/claude_desktop_config.json on Linux/Mac or %APPDATA%\Claude\claude_desktop_config.json on Windows):

{
  "mcpServers": {
    "oracle-cloud-mcp": {
      "command": "npx -y @oracle/mcp"
    }
  }
}

You'll need OCI credentials configured in your environment (typically via ~/.oci/config). Restart Claude Desktop and the server connects automatically.

Real-World Use Cases

Infrastructure as Conversation: "Create a new compute instance with 2 OCPUs in my production VCN, add it to the web-tier security group, and return the IP address"—executed entirely through Claude.
Database Diagnostics: Query your Autonomous Database for slow queries, check storage metrics, and generate a performance report without touching the console.
Compliance Automation: Pull audit logs for the last 30 days, filter for identity changes, and generate a summary report for your security team.

Full install guides for Claude Desktop, Cursor, Windsurf, and more at CuratedMCP.