OCI MCP Server

55 tools. 20 can modify or destroy data without limits.

6 destructive tools with no built-in limits. Policy required.

Last updated:

20 can modify or destroy data
35 read-only
55 tools total

Community server · catalogue entry verified 30/06/2026

How to control OCI MCP Server ↓

What OCI MCP Server exposes to your agents

Read (35) Write / Execute (14) Destructive / Financial (6)
Critical Risk

The most dangerous OCI MCP Server tools

20 of OCI MCP Server's 55 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

How to control OCI MCP Server

PolicyLayer is an MCP gateway — it sits between your AI agents and OCI MCP Server, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations
{
  "delete_bucket": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations
{
  "add_default_route": {
    "limits": [
      {
        "counter": "add_default_route_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "find_resources": {
    "limits": [
      {
        "counter": "find_resources_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

  1. Create a free account and register OCI MCP Server — nothing to install.
  2. Add these rules — paste them, or build them visually. Tune the limits to your setup.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
ENFORCE POLICY ON OCI →

Instant setup, no code required.

All 55 OCI MCP Server tools

READ 35 tools
Read find_resources Search ALL OCI resources. Examples: Read get_cluster Get OKE cluster. Read get_instance Get one instance. verbose=true returns full SDK object. Read get_namespace Tenancy's Object Storage namespace string. Read get_vault Get a vault. Read health_check Ping OCI: list regions, return count. Read list_alarms List monitoring alarms. Read list_applications List Functions applications. Read list_autonomous_databases List Autonomous Databases (ADW/ATP). Read list_availability_domains List availability domains (default compartment: tenancy). Read list_buckets List buckets. Read list_clusters List OKE clusters. Read list_compartments List compartments under a parent (default: tenancy). Read list_db_systems List Oracle Base Database systems. Read list_functions List functions in an application. Read list_images List platform images, newest first. Read list_instance_vnics List VNIC attachments (incl. public IP) for an instance. Read list_instances List compute instances. Read list_keys List keys in a vault. Get management_endpoint from get_vault. Read list_load_balancers List public/private Load Balancers. Read list_metrics List metric definitions in a compartment. namespace e.g. oci_computeagent. Read list_network_load_balancers List Network Load Balancers (L4). Read list_node_pools List OKE node pools. Read list_objects List objects in a bucket. Read list_regions List all OCI regions. Read list_resource_types List all resource types searchable via Resource Search. Read list_service_gateways List service gateways (private access to OCI services). Read list_services List OCI services available via service gateway. Read list_shapes List compute shapes available in a compartment. Read list_subnets List subnets, optionally filtered by VCN. Read list_users List IAM users. Read list_vaults List KMS vaults. Read list_vcns List VCNs. Read list_volumes List block volumes. Read whoami Tenancy/user/region from loaded OCI config — canonical 'connected?' check.

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

Questions about OCI MCP Server

Can an AI agent delete data through the OCI MCP Server MCP server? +

Yes. The OCI MCP Server server exposes 6 destructive tools including delete_bucket, delete_cluster, delete_load_balancer. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through OCI MCP Server? +

The OCI MCP Server server has 10 write tools including add_default_route, create_bucket, create_cluster. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach OCI MCP Server.

How many tools does the OCI MCP Server MCP server expose? +

55 tools across 4 categories: Destructive, Execute, Read, Write. 35 are read-only. 20 can modify, create, or delete data.

How do I enforce a policy on OCI MCP Server? +

Register the OCI MCP Server MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every OCI MCP Server tool call.

Deterministic rules across all 55 OCI MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

Instant setup, no code required.

55 OCI MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.