Available Tools

Once the MCP server is installed, your AI assistant can use the tools below. You don’t call these tools directly. Describe what you want in plain language, and the assistant chooses the right tools. See Example Prompts for ideas.

Inspect Your Workspace

These tools are read-only and never change your data.

Tool	Description
`debugInfo`	Checks your setup and returns the CLI version, data path, collections, and environments.
`listCollections`	Lists all collections with their folders and requests.
`listEnvironments`	Lists environment names and shows which one is active.
`getDetails`	Returns the full details of a collection, folder, or request by ID.

Create and Update

Tool	Description
`createCollection`	Creates a collection, optionally with a folder inside it.
`createFolder`	Adds a folder to a collection. Supports one level of subfolders.
`createRequest`	Creates a request from a curl command, with optional auth, tests, scripts, and docs.
`updateRequest`	Updates a request’s name, URL, method, headers, parameters, body, auth, tests, scripts, or docs.
`updateCollectionSettings`	Configures a collection or folder: base URL, shared auth, default headers, tests, scripts, environment, and run options. Can also rename it.

When updateRequest sets headers, parameters, or tests, it replaces the existing list instead of merging with it. The assistant includes every item that should remain.

Run Requests and Collections

These tools send real HTTP requests to your APIs.

Tool	Description
`runRequest`	Runs a single request and returns its status, response body, and test results.
`runCollection`	Runs a collection or folder and returns a per-request report with status, response time, and test results. Supports iterations, delay, concurrency, stop on failure, and a specific environment.

Both tools use the active environment unless you name a different one in your prompt.

Delete Requests

Tool	Description
`deleteRequest`	Permanently deletes a request from a collection or the Activity tab. This can’t be undone.

MCP Resources

The MCP server also provides two reference guides that the assistant reads on demand:

tc://docs: tool overview, URL and base URL conventions, the token capture pattern, and a scripting API reference.
tc://env-guide: guidance on using environment variables and handling secrets safely.

What the Assistant Can See

Data	Visible to the assistant
Collections, folders, and requests	Yes
Environment names and the active environment	Yes
Environment variable values	No. Only names are shared.
Saved auth values (tokens, passwords, secrets)	No. Values are masked.
Response bodies from `runRequest`	Yes. Responses longer than 50,000 characters are truncated.
Query parameter values in `runCollection` reports	No. Values are redacted.

The assistant can’t read or modify environment files directly. When a request needs a variable, the assistant tells you which variable to add to your active environment.

Limitations

The following must be done in the Thunder Client UI:

Environments: create, edit, and activate environments and their values.
AWS and NTLM auth: these credentials can’t be set through the MCP server.
Deleting collections and folders: only requests can be deleted through the MCP server.
Moving requests: moving a request to another collection or folder, or saving an Activity tab request to a collection.
Binary request bodies: file uploads using a binary body.
SSE, WebSocket, and gRPC: these request types can’t be created or run through the MCP server.