scaleway-mcp
Allows interaction with Scaleway's Cockpit observability service for monitoring and logging.
Enables management of Scaleway's managed Kubernetes (Kapsule) clusters, including listing, creating, and scaling.
Provides tools for managing Scaleway's Managed Redis instances, such as listing and configuring clusters.
Provides tools to manage Scaleway cloud infrastructure, including compute (Instances, Elastic Metal), containers (Kubernetes, Serverless), storage (Object, Block), databases (RDB, Redis), networking (VPC, Load Balancers, Public Gateway, DNS), and platform services (IAM, Secret Manager, Cockpit, Billing, Queues, Topics).
Integrates with Scaleway's Serverless Containers and Functions, allowing deployment and management of serverless workloads.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@scaleway-mcp list my instances"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
scaleway-mcp
A comprehensive Model Context Protocol (MCP) server for the Scaleway cloud platform. It exposes Scaleway's products as MCP tools so an AI agent (Claude Desktop, Claude Code, Cursor, and any other MCP client) can inspect and manage your cloud infrastructure.
21 products, 155+ tools spanning compute, containers, storage, databases, networking, messaging, and platform services — built on the official
scaleway-asyncSDK.Namespaced tools (
instances_list_servers,object_storage_list_buckets) so the agent can discover capabilities per product.Safe by default. Destructive operations are opt-in and, when enabled, only touch resources this server created.
Structured JSON output, not formatted text, so agents parse results reliably.
One small file per product, with a documented recipe for adding more.
Beta. The tool surface may change. Review the security model before pointing it at a production account.
Quick start
The server runs with uv / uvx and needs a
Scaleway API key.
Claude Code
claude mcp add scaleway \
-e SCW_ACCESS_KEY=SCWXXXXXXXXXXXXXXXXX \
-e SCW_SECRET_KEY=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
-e SCW_DEFAULT_PROJECT_ID=xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx \
-- uvx scaleway-mcpClaude Desktop / Cursor / other clients
Add to the client's MCP config (for Claude Desktop:
~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"scaleway": {
"command": "uvx",
"args": ["scaleway-mcp"],
"env": {
"SCW_ACCESS_KEY": "SCWXXXXXXXXXXXXXXXXX",
"SCW_SECRET_KEY": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
"SCW_DEFAULT_PROJECT_ID": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
}
}
}
}Full client instructions: docs/adding-to-your-agent.md.
Related MCP server: agentforge
Authentication
The server authenticates exactly like the scw CLI and the official SDK: it
reads ~/.config/scw/config.yaml when present and overlays the standard
environment variables (SCW_ACCESS_KEY, SCW_SECRET_KEY,
SCW_DEFAULT_PROJECT_ID, SCW_DEFAULT_REGION, SCW_DEFAULT_ZONE, ...). If you
already use the CLI, it works with no extra setup. See
docs/configuration.md.
Product coverage
Bundle | Products |
Compute & Containers | Instances, Elastic Metal, Kubernetes (Kapsule), Serverless Containers, Serverless Functions, Container Registry |
Storage & Databases | Object Storage (S3), Block Storage, Managed Databases (RDB), Redis |
Networking | VPC, Load Balancers, Public Gateway, Domains & DNS, Flexible IP |
Platform & Ops | IAM, Secret Manager, Cockpit (observability), Billing & Account |
Messaging & Queuing | Queues (SQS), Topics & Events (SNS) |
The complete, always-current tool list with parameters is in
docs/TOOLS.md. Ask the agent to call scaleway_list_products
to see what is enabled at runtime.
Enable only the products you need:
SCW_ENABLED_PRODUCTS=instance,object_storage,rdbSafety
Reads and creates by default; deletes are opt-in. Delete tools are not even registered unless
SCW_ENABLE_DELETE=true.Resources created through this server are tagged
created_by=scaleway-mcp. With deletes enabled, the server refuses to delete anything lacking that tag unlessSCW_ALLOW_DELETE_UNTAGGED=true.Use a least-privilege IAM key scoped to the products you enable. Details in docs/security.md.
Local development
git clone https://github.com/sndpl/scaleway-mcp
cd scaleway-mcp
uv sync --extra dev
uv run pytest # tests
uv run ruff check . # lint
uv run mypy # types
# Run against your account from a local checkout:
uvx --from . scaleway-mcpContributing
Adding a Scaleway product is a small, mechanical task. See CONTRIBUTING.md for the step-by-step recipe.
License
Apache-2.0. This is an unofficial, community project and is not affiliated with or endorsed by Scaleway.
Available Tools
129 toolsbaremetal_create_serverA
Create (order) an Elastic Metal server.
The server is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: name (required), offer_id (required), tags, zone, +3 more. Returns: object
baremetal_get_serverA
Get details of a single Elastic Metal server by ID.
Inputs: server_id (required), zone. Returns: object
baremetal_list_offersA
List available Elastic Metal offers (server hardware configurations).
Inputs: name, zone. Returns: result
baremetal_list_serversB
List Elastic Metal servers in a zone.
Inputs: name, tags, zone. Returns: result
baremetal_reboot_serverB
Reboot an Elastic Metal server.
Inputs: server_id (required), zone. Returns: object
baremetal_start_serverB
Start (power on) an Elastic Metal server.
Inputs: server_id (required), zone. Returns: object
baremetal_stop_serverB
Stop (power off) an Elastic Metal server.
Inputs: server_id (required), zone. Returns: object
billing_get_projectB
Get details of a single account project by ID.
Inputs: project_id. Returns: object
billing_list_consumptionsB
List current consumption (usage) records for an organization.
Inputs: project_id, category_name, billing_period, organization_id. Returns: result
billing_list_invoicesB
List invoices for an organization.
Inputs: organization_id. Returns: result
billing_list_projectsC
List account projects in an organization.
Inputs: name, organization_id. Returns: result
block_storage_create_snapshotA
Create a snapshot of a block volume.
The snapshot is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: volume_id (required), name, tags, zone, +1 more. Returns: object
block_storage_create_volumeA
Create an empty block volume.
The volume is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: size (required), name, tags, zone, +2 more. Returns: object
block_storage_get_volumeA
Get details of a single block volume by ID.
Inputs: volume_id (required), zone. Returns: object
block_storage_list_snapshotsB
List block snapshots in a zone.
Inputs: name, tags, zone, volume_id. Returns: result
block_storage_list_volumesC
List block volumes in a zone.
Inputs: name, tags, zone. Returns: result
cockpit_create_tokenB
Create a Cockpit token.
The secret key is only returned once, on creation.
Inputs: name (required), region, project_id, token_scopes. Returns: object
cockpit_get_configA
Get the Cockpit configuration (data source counts and retention) for a region.
Inputs: region. Returns: object
cockpit_get_current_planB
Get the current Cockpit pricing plan for a project.
Inputs: project_id. Returns: object
cockpit_list_data_sourcesA
List Cockpit data sources (metrics, logs, traces) in a region.
Inputs: region, project_id. Returns: result
cockpit_list_grafana_usersB
List Grafana users for a project.
Inputs: project_id. Returns: result
cockpit_list_tokensB
List Cockpit tokens in a region.
Inputs: region, project_id. Returns: result
containers_create_containerB
Create a container within a namespace.
The container is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: name (required), namespace_id (required), tags, region, +2 more. Returns: object
containers_create_namespaceA
Create a Serverless Containers namespace.
The namespace is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: name, tags, region, project, +1 more. Returns: object
containers_deploy_containerB
Deploy (start rolling out) a container.
Inputs: container_id (required), region. Returns: object
containers_get_containerA
Get details of a single container by ID.
Inputs: container_id (required), region. Returns: object
containers_get_namespaceB
Get details of a single namespace by ID.
Inputs: namespace_id (required), region. Returns: object
containers_list_containersB
List containers within a namespace.
Inputs: namespace_id (required), name, region. Returns: result
containers_list_namespacesC
List Serverless Containers namespaces in a region.
Inputs: name, region. Returns: result
databases_create_backupB
Create a manual backup of a database in a Managed Database instance.
Inputs: instance_id (required), database_name (required), name, region. Returns: object
databases_create_instanceB
Create a Managed Database instance.
The instance is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: engine (required), password (required), node_type (required), user_name (required), +6 more. Returns: object
databases_get_instanceA
Get details of a single Managed Database instance by ID.
Inputs: instance_id (required), region. Returns: object
databases_list_databasesA
List logical databases within a Managed Database instance.
Inputs: instance_id (required), region. Returns: result
databases_list_instancesB
List Managed Database instances in a region.
Inputs: name, tags, region. Returns: result
databases_list_usersB
List users of a Managed Database instance.
Inputs: instance_id (required), region. Returns: result
domains_list_dns_zone_recordsB
List the DNS records in a DNS zone.
This is a global service, so no region or zone applies.
Inputs: dns_zone (required), name. Returns: result
domains_list_dns_zonesA
List the DNS zones (subdomains) managed under a domain.
This is a global service, so no region or zone applies.
Inputs: domain (required). Returns: result
domains_list_domainsA
List the domain names registered in the account (registrar).
This is a global service, so no region or zone applies.
Inputs: domain. Returns: result
domains_update_dns_zone_recordsA
Apply a changeset to the records of a DNS zone.
Records have no tags, so nothing created here carries the managed marker.
Each entry in changes is a Scaleway record-change object with exactly one
of the keys add, set, delete or clear. For example::
[{"add": {"records": [
{"name": "www", "type": "A", "data": "1.2.3.4", "ttl": 3600}
]}}]Inputs: changes (required), dns_zone (required), return_all_records, disallow_new_zone_creation. Returns: object
flexible_ip_attach_flexible_ipB
Attach a flexible IP to an Elastic Metal server.
Inputs: fip_id (required), server_id (required), zone. Returns: object
flexible_ip_create_flexible_ipA
Create a flexible IP.
The IP is tagged created_by=scaleway-mcp so it can later be identified
and safely deleted by this server.
Inputs: description (required), tags, zone, is_ipv6, +3 more. Returns: object
flexible_ip_detach_flexible_ipC
Detach a flexible IP from its server.
Inputs: fip_id (required), zone. Returns: object
flexible_ip_get_flexible_ipB
Get details of a single flexible IP by ID.
Inputs: fip_id (required), zone. Returns: object
flexible_ip_list_flexible_ipsA
List flexible IPs in a zone.
Inputs: tags, zone. Returns: result
functions_create_namespaceB
Create a Serverless Functions namespace.
The namespace is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: name (required), tags, region, project_id, +1 more. Returns: object
functions_get_functionB
Get details of a single function by ID.
Inputs: function_id (required), region. Returns: object
functions_get_namespaceA
Get details of a single namespace by ID.
Inputs: namespace_id (required), region. Returns: object
functions_list_functionsB
List functions within a namespace.
Inputs: namespace_id (required), name, region. Returns: result
functions_list_namespacesB
List Serverless Functions namespaces in a region.
Inputs: name, region, project_id. Returns: result
functions_list_runtimesA
List function runtimes available in a region.
Inputs: region. Returns: result
iam_create_api_keyA
Create an IAM API key for an application or a user.
Sensitive and persistent: this mints long-lived credentials that outlive the
session and are not removed by disabling deletes. The secret key is returned
only once in the response. Only expose the iam product to trusted agents.
Exactly one of application_id or user_id identifies the bearer.
API keys carry no tags, so delete_api_key requires
SCW_ALLOW_DELETE_UNTAGGED=true to remove them.
Inputs: description (required), user_id, application_id, default_project_id. Returns: object
iam_create_applicationA
Create an IAM application.
Unlike compute resources, this server does not inject a managed tag onto
IAM applications; any tags you pass are stored verbatim. As a result,
delete_application cannot recognise applications as server-created and
requires SCW_ALLOW_DELETE_UNTAGGED=true to remove them.
Inputs: description (required), name, tags, organization_id. Returns: object
iam_list_api_keysC
List IAM API keys.
Inputs: user_id, application_id, organization_id. Returns: result
iam_list_applicationsB
List IAM applications in an organization.
Inputs: tag, name, organization_id. Returns: result
iam_list_groupsB
List IAM groups in an organization.
Inputs: tag, name, organization_id. Returns: result
iam_list_policiesB
List IAM policies in an organization.
Inputs: tag, policy_name, organization_id. Returns: result
iam_list_ssh_keysC
List IAM SSH keys.
Inputs: name, project_id, organization_id. Returns: result
iam_list_usersA
List IAM users in an organization.
Inputs: tag, organization_id. Returns: result
instances_create_serverB
Create a compute instance.
The instance is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: image (required), commercial_type (required), name, tags, +2 more. Returns: object
instances_get_serverB
Get details of a single compute instance by ID.
Inputs: server_id (required), zone. Returns: object
instances_list_imagesA
List instance images available in a zone.
Inputs: zone, public. Returns: result
instances_list_serversA
List compute instances (servers) in a zone.
Inputs: name, tags, zone. Returns: result
instances_list_volumesB
List instance volumes in a zone.
Inputs: zone. Returns: result
instances_perform_server_actionA
Perform a power or lifecycle action on an instance.
Inputs: action (required), server_id (required), zone. Returns: object
kubernetes_create_clusterA
Create a Kubernetes (Kapsule) cluster.
The cluster is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server. Pools are added separately
once the cluster exists.
Inputs: type_ (required), version (required), cni, name, +4 more. Returns: object
kubernetes_get_clusterA
Get details of a single Kubernetes cluster by ID.
Inputs: cluster_id (required), region. Returns: object
kubernetes_get_cluster_kubeconfigA
Get the kubeconfig file for a Kubernetes cluster.
Returns the Scaleway file descriptor whose content is the base64-encoded
kubeconfig YAML.
Sensitive: the kubeconfig contains cluster-admin credentials. Only expose
the k8s product to agents you trust with full control of the cluster.
Inputs: cluster_id (required), region. Returns: object
kubernetes_list_clustersB
List Kubernetes (Kapsule) clusters in a region.
Inputs: name, region. Returns: result
kubernetes_list_cluster_versionsA
List the Kubernetes versions available for new clusters in a region.
Inputs: region. Returns: result
kubernetes_list_poolsA
List the node pools of a Kubernetes cluster.
Inputs: cluster_id (required), name, region. Returns: result
load_balancers_create_lbB
Create a load balancer.
The load balancer is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: name (required), type_ (required), tags, zone, +2 more. Returns: object
load_balancers_get_lbA
Get details of a single load balancer by ID.
Inputs: lb_id (required), zone. Returns: object
load_balancers_list_backendsC
List the backends of a load balancer.
Inputs: lb_id (required), zone. Returns: result
load_balancers_list_frontendsB
List the frontends of a load balancer.
Inputs: lb_id (required), zone. Returns: result
load_balancers_list_ipsA
List load balancer flexible IPs in a zone.
Inputs: zone. Returns: result
load_balancers_list_lbsB
List load balancers in a zone.
Inputs: name, tags, zone. Returns: result
object_storage_create_bucketA
Create a bucket and tag it created_by=scaleway-mcp.
The managed tag lets this server safely delete the bucket later.
Inputs: bucket (required), region. Returns: object
object_storage_generate_presigned_urlA
Generate a temporary presigned GET URL for an object.
Inputs: key (required), bucket (required), region, expires_in. Returns: object
object_storage_get_object_metadataA
Get metadata for a single object (HEAD request).
Inputs: key (required), bucket (required), region. Returns: object
object_storage_list_bucketsA
List Object Storage buckets in a region.
Inputs: region. Returns: result
object_storage_list_objectsA
List objects in a bucket (one page).
S3 returns at most 1000 objects per page. When the result is truncated,
is_truncated is true and next_continuation_token can be passed back
in to fetch the next page.
Inputs: bucket (required), prefix, region, max_keys, +1 more. Returns: object
object_storage_put_objectB
Upload a text object to a bucket.
Inputs: key (required), bucket (required), content (required), region. Returns: object
public_gateway_create_gatewayB
Create a Public Gateway.
The gateway is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: type_ (required), name, tags, zone, +5 more. Returns: object
public_gateway_get_gatewayA
Get details of a single Public Gateway by ID.
Inputs: gateway_id (required), zone. Returns: object
public_gateway_list_gateway_networksB
List gateway-to-Private-Network attachments (gateway networks) in a zone.
Inputs: zone. Returns: result
public_gateway_list_gatewaysB
List Public Gateways in a zone.
Inputs: name, tags, zone. Returns: result
queues_activateA
Activate the Queues (SQS) service for a project and region.
Must be called once before the data-plane tools work.
Inputs: region, project_id. Returns: object
queues_create_credentialsA
Create SQS credentials for the data plane.
The returned secret_key is shown only once -- capture it now and set it
(with access_key) as SCW_SQS_SECRET_KEY / SCW_SQS_ACCESS_KEY for
the data-plane tools. MnQ resources are not taggable, so these are not tagged
as managed by this server.
Inputs: name, region, can_manage, project_id, +2 more. Returns: object
queues_create_queueC
Create a queue.
Inputs: name (required), region. Returns: object
queues_delete_messageA
Delete (acknowledge) a received message.
This is the normal end of the receive/process/delete cycle, not an
infrastructure delete, so it is not gated by SCW_ENABLE_DELETE.
Inputs: queue_name (required), receipt_handle (required), region. Returns: object
queues_get_infoB
Get the Queues (SQS) service status and endpoint for a project.
Inputs: region, project_id. Returns: object
queues_get_queue_urlB
Resolve a queue name to its URL.
Inputs: name (required), region. Returns: object
queues_list_credentialsA
List SQS credentials for a project (secret keys are not returned).
Inputs: region, project_id. Returns: result
queues_list_queuesB
List queue URLs in a region.
Inputs: region. Returns: object
queues_receive_messagesA
Receive messages from a queue.
Each returned message includes a ReceiptHandle needed to delete it after
processing (see delete_message).
Inputs: queue_name (required), region, max_messages, wait_time_seconds. Returns: object
queues_send_messageC
Send a message to a queue.
Inputs: body (required), queue_name (required), region. Returns: object
redis_create_clusterB
Create a Managed Redis cluster.
The cluster is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: name (required), version (required), password (required), node_type (required), +6 more. Returns: object
redis_get_clusterB
Get details of a single Managed Redis cluster by ID.
Inputs: cluster_id (required), zone. Returns: object
redis_list_clustersA
List Managed Redis clusters in a zone.
Inputs: name, tags, zone. Returns: result
redis_list_node_typesB
List Redis node types available in a zone.
Inputs: zone, include_disabled. Returns: result
registry_create_namespaceB
Create a container registry namespace.
The Container Registry API does not support resource-level tags, so unlike other products this namespace is not marked with the managed tag.
Inputs: name (required), region, project, is_public, +1 more. Returns: object
registry_get_imageB
Get details of a single container image by ID.
Inputs: image_id (required), region. Returns: object
registry_get_namespaceA
Get details of a single registry namespace by ID.
Inputs: namespace_id (required), region. Returns: object
registry_list_imagesB
List container images, optionally scoped to a namespace.
Inputs: name, region, namespace_id. Returns: result
registry_list_namespacesB
List container registry namespaces in a region.
Inputs: name, region. Returns: result
registry_list_tagsB
List the tags of a container image.
Inputs: image_id (required), name, region. Returns: result
scaleway_list_productsA
List the Scaleway products this server exposes and whether each is enabled.
Inputs: no named parameters. Returns: result
secrets_access_secret_versionA
Access the value stored in a secret's version.
WARNING: this returns SENSITIVE DATA. The data field of the result is
the base64-encoded secret payload. Handle and log the result with care.
Inputs: secret_id (required), region, revision. Returns: object
secrets_create_secretA
Create a secret.
The secret is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server. Creating a secret only
provisions the container; add a value with create_secret_version.
Inputs: name (required), tags, region, project, +1 more. Returns: object
secrets_create_secret_versionB
Add a new version (value) to an existing secret.
Inputs: data (required), secret_id (required), region, description. Returns: object
secrets_get_secretA
Get metadata of a single secret by ID.
This returns metadata only (name, tags, version count, ...), never the
stored secret value. Use access_secret_version to read a value.
Inputs: secret_id (required), region. Returns: object
secrets_list_secretsA
List secrets in a region.
Only secrets that are not scheduled for deletion are returned.
Inputs: name, tags, region. Returns: result
secrets_list_secret_versionsA
List the versions of a secret.
Returns version metadata (revision, status, timestamps) only, never the stored values.
Inputs: secret_id (required), region. Returns: result
topics_activateA
Activate the Topics & Events (SNS) service for a project and region.
Must be called once before the data-plane tools work.
Inputs: region, project_id. Returns: object
topics_create_credentialsA
Create SNS credentials for the data plane.
The returned secret_key is shown only once -- capture it now and set it
(with access_key) as SCW_SNS_SECRET_KEY / SCW_SNS_ACCESS_KEY for
the data-plane tools. MnQ resources are not taggable, so these are not tagged
as managed by this server.
Inputs: name, region, can_manage, project_id, +2 more. Returns: object
topics_create_topicC
Create a topic.
Inputs: name (required), region. Returns: object
topics_get_infoA
Get the Topics & Events (SNS) service status and endpoint for a project.
Inputs: region, project_id. Returns: object
topics_list_credentialsB
List SNS credentials for a project (secret keys are not returned).
Inputs: region, project_id. Returns: result
topics_list_subscriptionsC
List the subscriptions of a topic.
Inputs: topic_arn (required), region. Returns: object
topics_list_topicsA
List topic ARNs in a region.
Inputs: region. Returns: object
topics_publishA
Publish a message to a topic.
Note: SNS silently drops messages published to a topic with no subscribers.
Inputs: message (required), topic_arn (required), region. Returns: object
topics_subscribeC
Subscribe an endpoint to a topic.
Inputs: endpoint (required), protocol (required), topic_arn (required), region. Returns: object
topics_unsubscribeB
Remove a subscription.
Inputs: subscription_arn (required), region. Returns: object
vpc_create_private_networkB
Create a Private Network.
The Private Network is tagged created_by=scaleway-mcp so it can later be
identified and safely deleted by this server.
Inputs: name, tags, region, vpc_id, +3 more. Returns: object
vpc_create_vpcA
Create a Virtual Private Cloud.
The VPC is tagged created_by=scaleway-mcp so it can later be identified
and safely deleted by this server.
Inputs: name, tags, region, project, +1 more. Returns: object
vpc_get_private_networkA
Get details of a single Private Network by ID.
Inputs: private_network_id (required), region. Returns: object
vpc_get_vpcA
Get details of a single Virtual Private Cloud by ID.
Inputs: vpc_id (required), region. Returns: object
vpc_list_private_networksB
List Private Networks in a region.
Inputs: name, tags, region, vpc_id. Returns: result
vpc_list_vpcsC
List Virtual Private Clouds in a region.
Inputs: name, tags, region. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 129 tools
Each tool is clearly scoped to a specific service and action (e.g., baremetal_, instances_, etc.). Within each service, tools like create, get, list, and update are distinct. No two tools could easily be confused for the same operation.
All tools follow a consistent {service}_{verb}_{resource} pattern using snake_case. For example, baremetal_create_server, block_storage_list_volumes. No mixing of styles or inconsistent abbreviations.
129 tools is large but appropriate for a comprehensive cloud MCP covering many services. Each service has a focused set (3-12 tools), so the count reflects the breadth of Scaleway's offerings without feeling bloated for its scope.
Many services lack critical lifecycle operations, notably delete and update tools. For instance, there is no tool to delete a baremetal server, compute instance, Kubernetes cluster, or block volume. Additionally, the functions service has no create_function tool. The toolkit covers basic read and create operations but misses essential management actions.
Maintenance
Related MCP Connectors
214 tools for AI agents over one MCP endpoint: image, video, audio, 3D, data, storage, mail, GPUs.
MCP server for AI agents to plan, verify, and deploy Cloudflare-native apps.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
An MCP gateway for teams deploying AI tools at scale to employee-facing and headless agents.
1
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceA comprehensive MCP server providing over 390 tools across 66 providers, including major SaaS platforms like GitHub, Slack, and Stripe. It enables AI assistants to interact directly with a wide array of public APIs and utility services through a single interface.-
- AlicenseAqualityDmaintenanceMCP server that exposes 300+ AI agents as tools via a single API key. Supports listing agents, invoking any agent with chat-completion style messages, checking agent health, and retrieving platform statistics.54MIT
- FlicenseCqualityDmaintenanceComprehensive MCP server for managing all IBM Cloud services, offering 180+ tools across 16 domains to enable AI assistants to fully manage IBM Cloud infrastructure.1002-
- AlicenseBqualityDmaintenanceEnables conversational management of Hetzner Cloud, DNS, and Storage Box infrastructure through 137 MCP tools for CRUD operations and actions like server power control, DNS zone management, and storage box snapshots.100MIT