Provider REST API

15 actions

Connect Claude to Cloudinary

Search, upload, tag, organize and delete images and videos in a Cloudinary product environment, and read its usage. Toolspoke puts 15 of its actions behind one MCP endpoint that Claude, Cursor, and Codex all speak.

Connect View pricing

Connection: Provider REST API
Authentication: API key and secret
Actions exposed: 15
What a call bills: Nothing
Adapter: Maintained by Toolspoke

Connected in three steps

1
Install Cloudinary
Open the marketplace in your workspace, add Cloudinary to the project your agents work in, and it appears on the gateway immediately.
2
Connect the credential
Authenticate with api key and secret. Where to get one, and what it has to be able to reach, is the next section.
3
Point your agent at the gateway
Give your client one address, https://toolspoke.com/mcp. Claude Code takes it as a command, Claude and Claude Desktop add it as a custom connector, and Cursor, Codex and VS Code each read it from a config file of their own.

.mcp.json

{
  "mcpServers": {
    "toolspoke": {
      "type": "http",
      "url": "https://toolspoke.com/mcp"
    }
  }
}

One block covers every tool you have installed. Cloudinary shows up in the client as soon as your policy allows it, and so does everything else you install later.

Where the address goes, per client

Claude Code: Run it in your project, then /mcp to sign in
claude mcp add --transport http toolspoke https://toolspoke.com/mcp
Claude and Claude Desktop: Settings, then Connectors, then Add custom connector
https://toolspoke.com/mcp
Cursor: ~/.cursor/mcp.json, or .cursor/mcp.json for one project
{ "mcpServers": { "toolspoke": { "url": "https://toolspoke.com/mcp" } } }
Codex: ~/.codex/config.toml
[mcp_servers.toolspoke] url = "https://toolspoke.com/mcp"
VS Code: .vscode/mcp.json, or the MCP: Add Server command
{ "servers": { "toolspoke": { "type": "http", "url": "https://toolspoke.com/mcp" } } }

What Cloudinary asks for

API key and secret. You provide it once, when you install the connector. Toolspoke encrypts it at rest and decrypts it only for the length of a single call, and the gateway attaches it to the outbound request itself, so it is never part of the arguments an agent sends.

Cloud nameRequired: Cloudinary Console → Settings → API Keys. The cloud name is shown at the top of that page.; demo
API key and API secretRequired: Cloudinary Console → Settings → API Keys → Generate New API Key. Paste the API key, a colon, then the API secret. It is sent as HTTP Basic. A key created with a restricted role is limited to that role's permissions.; 123456789012345:abcdEFGHijklMNOPqrstUVWXyz0

What Claude can do in Cloudinary

15 actions, each one declared and named by the connector rather than discovered at runtime. A workspace policy grants a person all of them, a hand-picked selection, everything on the read side, everything on the write side, or none.

Reads: 9Reads
Writes: 4Writes
Destructive: 2Destructive

Reads

Fetches data and changes nothing.

get_usage
Report the product environment's plan, credits, storage, bandwidth, transformations, request counts and asset counts. It takes no arguments and needs a working key and the right cloud name, which is why an install tests itself with it.
list_resources
List assets of one resource type and delivery type, newest first, optionally only those whose public ID starts with a prefix or that were created after a date. Paged by next_cursor.
get_resource
Read one asset by public ID - format, size, dimensions, URLs, tags, context, structured metadata and derived versions.
search_assets
Search every asset with a Lucene-like expression over public_id, folder, asset_folder, tags, context, format, bytes, width, created_at and more - e.g. 'tags:shoe AND format:png', 'uploaded_at>1w', 'asset_folder="Marketing"'. Paged by next_cursor.
list_resources_by_tag
List the assets of one resource type that carry a given tag.
list_resources_in_folder
List the assets placed in one asset folder (dynamic folder mode), of every resource type.
list_root_folders
List the top-level folders of the product environment.
list_subfolders
List the folders directly inside a folder.
list_tags
List the tags in use on assets of one resource type, optionally only those starting with a prefix.

Writes

Creates or updates something on the other side.

upload_from_url
Upload an asset from a public HTTPS URL (or a data URI). It never overwrites an existing asset with the same public ID - that is fixed to false here, so an upload cannot replace something already in use.
update_resource
Update one asset's tags, contextual metadata, structured metadata, display name or folder. tags replaces the asset's whole tag list. context and metadata are pipe-separated key=value pairs, e.g. alt=Red shoe|caption=Sale.
rename_resource
Change an asset's public ID, which changes its delivery URL - anything linking to the old URL breaks. It never overwrites an asset already at the new ID; that is fixed to false here.
create_folder
Create an empty folder, including any missing parents in the path.

Destructive

Deletes or permanently alters something. Worth granting on purpose.

delete_resources
Delete up to 100 assets by public ID, with their derived versions. Any page still using their URLs breaks; recovery is only possible if backups are enabled on the account.
delete_folder
Delete an empty folder. Cloudinary refuses a folder that still holds assets, so delete or move them first.

What it will not do

Enforced by the gateway rather than left to convention, which is why each of these can be stated flatly.

It cannot call anything else: The 15 actions above are the whole of it. A call to any other name is refused before it reaches Cloudinary rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time.
It reaches no further than your credential: Toolspoke holds no access to Cloudinary of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either.
It never hears from Cloudinary: Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in Cloudinary. An agent has to ask.
It does not smooth over provider limits: Toolspoke does not retry, queue or back off around Cloudinary's own rate limits. A call that Cloudinary refuses comes back to the agent as a failed call.

Before you connect it

What can Claude do in Cloudinary?

15 named actions: 9 that only read, 4 that write and 2 that delete or permanently alter something. They include get_usage, list_resources and get_resource. Nothing outside that list is reachable: the connector declares each operation by name rather than proxying whatever an agent asks for.

What credentials does the Cloudinary connector need?

API key and secret. The connector asks for cloud name and api key and api secret. Values are encrypted at rest and attached to the outbound request by the gateway, so they are never part of the arguments an agent sends and never reach the audit log.

Does the Cloudinary connector work with Cursor and Codex, or only Claude?

Any client that speaks MCP, and every one of them gets the same 15 actions. There is a single address, https://toolspoke.com/mcp. Claude Code adds it with claude mcp add --transport http, Claude and Claude Desktop take it as a custom connector in settings, Cursor reads it from .cursor/mcp.json, Codex from ~/.codex/config.toml, and VS Code from .vscode/mcp.json. Each of them signs in to the gateway itself, so there is no key to paste.

What does the Cloudinary connector not do?

The 15 actions above are the whole of it. A call to any other name is refused before it reaches Cloudinary rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time. Toolspoke holds no access to Cloudinary of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either. Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in Cloudinary. An agent has to ask. Toolspoke does not retry, queue or back off around Cloudinary's own rate limits. A call that Cloudinary refuses comes back to the agent as a failed call.

Can I limit which actions an agent can call?

Yes, in two places. The project switches Cloudinary's actions on and off one at a time, for everyone in the project at once, and the screen groups them by read, write and destructive so turning off everything that deletes is one click. An individual agent key can then be narrowed further, to particular toolkits in a project and to particular actions in a toolkit. Whatever it was granted, a key never reaches a project its owner cannot.

What gets recorded when an agent calls Cloudinary?

Every attempt, with the agent that made it and the person that agent belongs to, the full request payload, the response payload, the status, the duration, and what the call cost in sandbox time and model tokens. Values whose key names a secret are masked out before the record is shown to anyone. An operation the connector marks as not retained never has its response body written at all, so the gateway keeps no second copy of what was read.

More design connectors

Browse all of them