CP MCP Server

URL: https://mcp.computeportal.io/mcp

The Compute Portal MCP server exposes all CP platform capabilities as tools that AI assistants (Claude, Cursor, Windsurf, and any MCP-compatible client) can call directly. This lets you manage GPU workloads, deploy services, run sandboxes, provision databases, inspect billing, and more through natural language — without leaving your AI assistant.

---

Protocol

The CP MCP server implements the Model Context Protocol (MCP) using the Streamable HTTP transport.

Property	Value
Protocol	MCP (Model Context Protocol)
Transport	Streamable HTTP
Endpoint	https://mcp.computeportal.io/mcp
SDK	@modelcontextprotocol/sdk
Auth	CP_API_TOKEN environment variable
Total tools	21

Authentication

The MCP server authenticates using your CP API token:

CP_API_TOKEN=cp_live_<hex-string>

The token format is cp_live_ followed by a lowercase hex string — the same token used for direct API and CLI calls. Find yours at https://dashboard.computeportal.io under Settings → API Tokens.

---

Integrations

Claude Desktop

Add to your Claude Desktop config file.

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "computeportal": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.computeportal.io/mcp"],
      "env": {
        "CP_API_TOKEN": "cp_live_4a7f3b9c2e1d8a06"
      }
    }
  }
}

Restart Claude Desktop after saving. A hammer icon in the chat input area confirms MCP tools are active.

Verify: Ask Claude "List my CP services" — it should call cp_service_list and return your running services.

---

Cursor

Open Cursor Settings → MCP and add:

{
  "computeportal": {
    "command": "npx",
    "args": ["-y", "mcp-remote", "https://mcp.computeportal.io/mcp"],
    "env": {
      "CP_API_TOKEN": "cp_live_4a7f3b9c2e1d8a06"
    }
  }
}

If your Cursor version supports HTTP MCP natively:

{
  "computeportal": {
    "url": "https://mcp.computeportal.io/mcp",
    "headers": {
      "Authorization": "Bearer cp_live_4a7f3b9c2e1d8a06"
    }
  }
}

---

Windsurf

In ~/.windsurf/mcp.json:

{
  "servers": {
    "computeportal": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.computeportal.io/mcp"],
      "env": {
        "CP_API_TOKEN": "cp_live_4a7f3b9c2e1d8a06"
      }
    }
  }
}

---

Direct HTTP

For clients that support Streamable HTTP natively:

URL:     https://mcp.computeportal.io/mcp
Headers: Authorization: Bearer cp_live_4a7f3b9c2e1d8a06
         Content-Type: application/json

MCP initialize handshake:

curl -X POST https://mcp.computeportal.io/mcp \
  -H "Authorization: Bearer cp_live_4a7f3b9c2e1d8a06" \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "initialize",
    "params": {
      "protocolVersion": "2024-11-05",
      "capabilities": {},
      "clientInfo": { "name": "my-client", "version": "1.0.0" }
    }
  }'

---

Tools Reference

The CP MCP server exposes 21 tools grouped by domain.

Service Tools

cp_deploy

Deploy a new service from a Docker image or a public GitHub repo. When repo is provided, CP clones and builds the image internally — no registry setup needed. Build operations time out after 10 minutes.

Inputs

Parameter	Type	Required	Description
image	string	No*	Docker image URI (e.g. myorg/agent:latest)
repo	string	No*	Public GitHub repo URL — CP builds from the Dockerfile
name	string	No	Service name (defaults to derived from image or repo)
region	string	No	Region ID (e.g. eu-west)
gpu	string	No	GPU type: rtx4090 | rtx5090
gpu_count	integer	No	Number of GPUs (min: 1)
env	object	No	Environment variables as key-value pairs
domains	array	No	Custom hostnames to attach after deploy

* Exactly one of image or repo is required.

Example — deploy from Docker image

"Deploy a new service called ml-inference using the image myorg/model:v2"

{
  "tool": "cp_deploy",
  "arguments": {
    "name": "ml-inference",
    "image": "myorg/model:v2"
  }
}

Example — deploy from GitHub repo

"Deploy https://github.com/my-org/my-app as my-app"

{
  "tool": "cp_deploy",
  "arguments": {
    "name": "my-app",
    "repo": "https://github.com/my-org/my-app"
  }
}

Returns

{
  "service_id": "my-app",
  "name": "my-app",
  "url": "https://my-app.cpctl.app",
  "status": "running"
}

---

cp_service_list

List all running services. Optionally filter by region.

Inputs

Parameter	Type	Required	Description
region	string	No	Filter by region ID

Example prompt

"Show me all my running services"

---

cp_service_status

Get current status, replica count, and health metrics for a service.

Inputs

Parameter	Type	Required	Description
name	string	Yes	Service name

Example prompt

"What's the status of ml-inference?"

---

cp_service_logs

Fetch the most recent log lines from a running service.

Inputs

Parameter	Type	Required	Description
name	string	Yes	Service name
tail	integer	No	Number of log lines to return (1–5000, default: 100)

Example prompt

"Show me the last 200 lines of logs from ml-inference"

{
  "tool": "cp_service_logs",
  "arguments": {
    "name": "ml-inference",
    "tail": 200
  }
}

---

cp_scale

Scale replica count or update CPU and memory limits for a service.

Inputs

Parameter	Type	Required	Description
name	string	Yes	Service name
count	integer	No	Number of replicas (min: 0)
cpu	string	No	CPU limit in vCPUs (e.g. '2')
ram	string	No	Memory limit (e.g. '4Gi')

At least one of count, cpu, or ram must be provided.

Example prompt

"Scale ml-inference to 3 replicas with 4Gi memory"

{
  "tool": "cp_scale",
  "arguments": {
    "name": "ml-inference",
    "count": 3,
    "ram": "4Gi"
  }
}

---

cp_destroy

Permanently destroy a service and release all associated resources. This action is irreversible.

Inputs

Parameter	Type	Required	Description
name	string	Yes	Service name to destroy

Returns { "destroyed": "ml-inference" }

---

cp_env_set

Set one or more environment variables on a running service. Triggers a rolling restart to apply changes.

Inputs

Parameter	Type	Required	Description
service	string	Yes	Service name
vars	object	Yes	Key-value pairs to set

Example prompt

"Set MODEL_PATH=/models/v2 and BATCH_SIZE=32 on ml-inference"

{
  "tool": "cp_env_set",
  "arguments": {
    "service": "ml-inference",
    "vars": {
      "MODEL_PATH": "/models/v2",
      "BATCH_SIZE": "32"
    }
  }
}

Returns { "service": "ml-inference", "set": 2 }

---

GPU Tools

cp_gpu_list

List available GPU nodes with specs, live utilization, VRAM usage, and temperature.

Inputs

Parameter	Type	Required	Description
region	string	No	Filter by region
type	string	No	Filter by GPU type: rtx4090 | rtx5090

Example prompt

"What RTX 4090 nodes are available in eu-west?"

---

cp_gpu_reserve

Reserve GPU capacity for a fixed duration. Returns a reservation ID and hourly rate.

Inputs

Parameter	Type	Required	Description
type	string	Yes	GPU type: rtx4090 | rtx5090
count	integer	No	Number of GPUs (min: 1, default: 1)
duration	string	No	Reservation duration (e.g. 4h, 24h, 7d, default: 1h)
region	string	No	Region (defaults to account default)

Example prompt

"Reserve 2 RTX 4090 GPUs in eu-west for 8 hours"

{
  "tool": "cp_gpu_reserve",
  "arguments": {
    "type": "rtx4090",
    "count": 2,
    "duration": "8h",
    "region": "eu-west"
  }
}

Returns Reservation object with ID, expiry time, and hourly rate.

---

cp_gpu_run

Submit a one-shot GPU job. Billed per second. Returns a job ID and initial status immediately.

Inputs

Parameter	Type	Required	Description
image	string	Yes	Docker image to run
script	string	No	Command or script to execute inside the container
gpu	string	No	GPU type (default: rtx4090)
gpu_count	integer	No	Number of GPUs (min: 1, default: 1)
region	string	No	Region ID

Example prompt

"Run a one-shot job with image myorg/trainer:latest on an RTX 4090"

---

Node Tools

cp_node_list

List all bare-metal nodes with hardware specs (CPU cores, RAM, GPU type/count) and reservation status.

Inputs

Parameter	Type	Required	Description
region	string	No	Filter by region
gpu	string	No	Filter by GPU type: rtx4090 | rtx5090

Example prompt

"Show me all available bare-metal RTX 5090 nodes"

---

Billing Tools

cp_billing_balance

Return current account balance, credit, hourly burn rate, and alert threshold.

Inputs — None

Example prompt

"What's my current balance and burn rate?"

Returns

{
  "balance": 142.50,
  "credit": 0,
  "burn_rate_hourly": 3.20,
  "alert_threshold": 20.00
}

---

cp_billing_usage

Return itemized billing usage for a given period, optionally filtered by service.

Inputs

Parameter	Type	Required	Description
period	string	No	today | month | YYYY-MM (default: today)
service	string	No	Filter by service name

Example prompt

"Show me this month's billing breakdown by service"

---

Region Tools

cp_region_list

List all Compute Portal regions with compliance certifications, GPU availability, and latency.

Inputs — None

Example prompt

"What regions are available and which have RTX 5090s?"

---

Database Tools

cp_db_create

Provision a managed database. Returns connection credentials.

Inputs

Parameter	Type	Required	Description
type	string	Yes	Database engine: postgres | redis | mysql | mongo
name	string	No	Database name (defaults to {type}-db)
region	string	No	Region (defaults to account default)
size_gb	integer	No	Storage in GB (min: 1, default: 10)

Example prompt

"Provision a 20GB PostgreSQL database called my-app-db in eu-west"

{
  "tool": "cp_db_create",
  "arguments": {
    "type": "postgres",
    "name": "my-app-db",
    "region": "eu-west",
    "size_gb": 20
  }
}

Returns Database host, port, and credentials.

---

Sandbox Tools

Sandboxes are isolated GPU-attached microVMs for AI agent code execution. They expose SSH access and a command execution API.

cp_sandbox_create

Spin up a GPU-attached sandbox microVM.

Inputs

Parameter	Type	Required	Description
image	string	Yes	Container image
gpu	string	No	GPU type (default: rtx4090)
gpu_count	integer	No	Number of GPUs (min: 1, default: 1)
region	string	No	Region ID
timeout	string	No	Sandbox lifetime (default: 1h, e.g. 2h, 30m)

Example prompt

"Create a GPU sandbox using image myorg/dev-env:latest with 2 RTX 4090s for 2 hours"

Returns Sandbox ID, SSH connection details, and GPU info.

---

cp_sandbox_exec

Execute a command inside a running sandbox and return stdout, stderr, exit code, and duration.

Inputs

Parameter	Type	Required	Description
id	string	Yes	Sandbox ID
command	array of strings	Yes	Command and arguments (e.g. ["python", "train.py", "--epochs", "10"])

Example prompt

"Run python train.py --epochs 10 in sandbox sb_abc123"

{
  "tool": "cp_sandbox_exec",
  "arguments": {
    "id": "sb_abc123",
    "command": ["python", "train.py", "--epochs", "10"]
  }
}

Returns

{
  "stdout": "Epoch 10/10 — loss: 0.042",
  "stderr": "",
  "exit_code": 0,
  "duration_ms": 18420
}

---

cp_sandbox_destroy

Destroy a sandbox and release its GPU resources. This action is irreversible.

Inputs

Parameter	Type	Required	Description
id	string	Yes	Sandbox ID

Returns { "destroyed": "sb_abc123" }

---

Domain Tools

cp_domain_add

Attach a custom hostname to a CP service. Validates the DNS CNAME record and provisions TLS automatically. The CNAME must point to cpctl.app.

Inputs

Parameter	Type	Required	Description
service	string	Yes	Service name (e.g. my-app)
hostname	string	Yes	Custom hostname (e.g. dashboard.myapp.com)
skip_dns	boolean	No	Skip DNS CNAME validation (default: false)

Example prompt

"Attach api.myapp.com to my ml-inference service — I've already set up the CNAME"

{
  "tool": "cp_domain_add",
  "arguments": {
    "service": "ml-inference",
    "hostname": "api.myapp.com"
  }
}

Returns

{
  "hostname": "api.myapp.com",
  "status": "active",
  "service": "ml-inference",
  "created_at": "2026-08-29T10:52:36.153Z"
}

---

cp_domain_list

List all domains attached to a service — both the default {service}.cpctl.app and any custom hostnames.

Inputs

Parameter	Type	Required	Description
service	string	Yes	Service name

Example prompt

"What domains are attached to ml-inference?"

---

cp_domain_remove

Remove a custom hostname from a CP service. The default {service}.cpctl.app domain cannot be removed.

Inputs

Parameter	Type	Required	Description
service	string	Yes	Service name
hostname	string	Yes	Custom hostname to remove

Returns { "removed": "api.myapp.com", "service": "ml-inference" }

---

Usage Patterns

The AI assistant maps natural language to tool calls. You do not need to know tool names — just describe what you want.

Service management

"Show me all my running services"
→ cp_service_list

"Deploy https://github.com/my-org/trainer as a service called model-trainer with 2 RTX 4090s"
→ cp_deploy with repo + gpu + gpu_count

"Scale ml-inference to 3 replicas with 8Gi RAM"
→ cp_scale

"Delete the ml-inference service"
→ cp_destroy

GPU workloads

"What RTX 4090 nodes are available?"
→ cp_gpu_list

"Reserve 4 RTX 4090s for 24 hours"
→ cp_gpu_reserve

"Run a one-shot training job with image myorg/trainer:latest"
→ cp_gpu_run

Sandboxes

"Create a GPU sandbox for 2 hours using image myorg/dev-env:latest"
→ cp_sandbox_create

"Run nvidia-smi in sandbox sb_abc123"
→ cp_sandbox_exec

Databases

"Provision a 20GB PostgreSQL database called my-app-db"
→ cp_db_create

Billing

"What's my current balance and hourly burn rate?"
→ cp_billing_balance

"Show me this month's usage by service"
→ cp_billing_usage

Domains

"Attach api.myapp.com to ml-inference — CNAME is already set"
→ cp_domain_add

Debugging

"Show me the last 200 lines of logs from ml-inference"
→ cp_service_logs

"My service is crashing — show me its status and recent error logs"
→ cp_service_status then cp_service_logs

---

Troubleshooting

"No tools available" in Claude Desktop

1. Verify claude_desktop_config.json is valid JSON (no trailing commas)
2. Confirm CP_API_TOKEN starts with cp_live_
3. Fully quit and relaunch Claude Desktop (Cmd+Q on macOS, not just close the window)
4. Check npx is in your PATH: which npx

"Unauthorized" errors when calling tools

Your token may be invalid or expired. Generate a new one at https://dashboard.computeportal.io under Settings → API Tokens, then update your MCP config.

DNS validation failure when adding a domain

The cp_domain_add tool validates that the hostname's CNAME resolves to cpctl.app. If validation fails:

1. Confirm the DNS record is saved and set to DNS only (not proxied through Cloudflare)
2. Wait for propagation — typically 1–5 minutes for Cloudflare, up to 48 hours for other registrars
3. Verify: dig CNAME dashboard.myapp.com — the answer should point to cpctl.app
4. Pass skip_dns: true to attach the domain immediately without DNS validation

Deploy from repo is slow

Repo-based deploys involve cloning, building, and pushing a Docker image. This typically takes 1–5 minutes depending on repo size and Dockerfile complexity. The tool has a 10-minute timeout — if exceeded, check the CP dashboard for build status.

Tool calls time out

Most tool calls complete in under 5 seconds. Operations that involve provisioning (databases, sandboxes, reservations) may take 15–30 seconds. If a call times out in your client, check the CP Dashboard to confirm whether the operation completed.

---

Security

• Tokens are scoped to your CP account. Never share your cp_live_xxx token.
• The MCP server does not store tokens — it passes them to the CP API on each call.
• All traffic is HTTPS. Plain HTTP is not supported.
• Revoke tokens at any time from the CP Dashboard.

---

Quick Reference

Tool	Description
cp_deploy	Deploy from image or public GitHub repo
cp_service_list	List all running services
cp_service_status	Status, replicas, and health for a service
cp_service_logs	Fetch recent log lines
cp_scale	Scale replicas, CPU, or memory
cp_destroy	Permanently destroy a service
cp_env_set	Set environment variables on a service
cp_gpu_list	List GPU nodes with utilization and specs
cp_gpu_reserve	Reserve GPU capacity for a fixed duration
cp_gpu_run	Submit a one-shot GPU job
cp_node_list	List bare-metal nodes with hardware specs
cp_billing_balance	Current balance, credit, and burn rate
cp_billing_usage	Itemized usage for a period
cp_region_list	Available regions with GPU availability
cp_db_create	Provision a managed database
cp_sandbox_create	Create a GPU-attached microVM sandbox
cp_sandbox_exec	Execute a command inside a sandbox
cp_sandbox_destroy	Destroy a sandbox
cp_domain_add	Attach a custom hostname to a service
cp_domain_list	List domains on a service
cp_domain_remove	Remove a custom hostname

---

Related

• Quickstart Guide — get a token and make your first API call
• cpctl CLI — terminal-based equivalent of these tools
• CP API Reference — full HTTP API documentation
• MCP Protocol Specification — official MCP docs