cloudcraft-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CLOUDCRAFT_API_KEY | Yes | API key (Bearer). Generated in Cloudcraft User settings. | |
| CLOUDCRAFT_BASE_URL | No | Override for proxies or future API versions. | https://api.cloudcraft.co |
| CLOUDCRAFT_LOG_LEVEL | No | Stderr log verbosity (DEBUG / INFO / WARNING / ERROR). | WARNING |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| whoamiA | Return the Cloudcraft user profile for the current API key. Useful as a sanity check after configuring a new key. |
| list_blueprintsA | List every Cloudcraft blueprint (diagram) in the authenticated account. Returns a compact summary per blueprint (id, name, tags, updatedAt).
Use :func: |
| get_blueprintA | Fetch the full blueprint payload (nodes, edges, groups, layout). Args:
blueprint_id: Blueprint UUID from :func: |
| create_blueprintB | Create a new Cloudcraft blueprint. Args: name: Display name for the new blueprint. data: Full blueprint payload (nodes, edges, groups, surfaces, text, icons, connectors, theme, projection, etc.). See README for a minimal example. Returns the created blueprint metadata including the assigned id. |
| update_blueprintA | Replace the full data payload of an existing blueprint. Args:
blueprint_id: Blueprint UUID.
data: Full blueprint payload (same shape as :func: |
| delete_blueprintA | Delete a Cloudcraft blueprint after exact-id confirmation. Args: blueprint_id: Blueprint UUID. confirm_blueprint_id: Repeat the exact UUID to acknowledge the irreversible operation. |
| export_blueprint_imageA | Render a Cloudcraft blueprint to an image file on disk. Writes are restricted to Args:
blueprint_id: Blueprint UUID.
format: One of Returns:
|
| export_blueprint_budgetA | Export a blueprint's cost estimate (budget) to a CSV or XLSX file on disk. Same write sandbox as :func: Args:
blueprint_id: Blueprint UUID.
format: Returns:
|
| list_teamsB | List the teams linked to the Cloudcraft account (id, name, members). |
| list_aws_accountsA | List AWS accounts registered with Cloudcraft for live-scan snapshots. |
| snapshot_awsB | Scan one AWS region into a blueprint JSON via Cloudcraft live-scan. |
| list_azure_accountsA | List Azure accounts registered with Cloudcraft for live-scan snapshots. |
| snapshot_azureC | Scan one Azure region into a blueprint JSON via Cloudcraft live-scan. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 13 tools
Each tool maps to a distinct resource/action pair: team/profile read, blueprint CRUD, two clearly separated export formats (image vs budget), and account listing vs region snapshots. The only near-overlap, export_blueprint_image vs export_blueprint_budget, is made unambiguous by explicit format and output descriptions.
Almost everything follows a clean verb_noun snake_case pattern (list_*, get_*, create_*, update_*, delete_*, export_*, snapshot_*). The lone deviation is `whoami`, which breaks the pattern but is idiomatic and unambiguous.
13 tools is well within the ideal 3-15 range and each earns its place: 5 blueprint CRUD tools, 2 exports, 2 account listers, 2 snapshotters, plus team/profile reads. No filler tools.
Blueprint lifecycle is fully covered (list/get/create/update/delete plus image and budget export), and both AWS and Azure snapshot paths exist. Minor gaps: no account-level get/registration tools, no team management, and update_blueprint replaces the whole payload so rename/metadata edits are awkward — all workaroundable.