Connect ElevenLabs to Cursor

Speak text in any voice, and generate images and video from a prompt. Add Toolspoke to Cursor once and it can use 18 ElevenLabs actions. Your ElevenLabs credential stays in Toolspoke, and you choose which actions Cursor may call.

Set it up in Cursor

Three steps, the first one once for every toolkit you will ever add.

  1. Add Toolspoke to Cursor

    ~/.cursor/mcp.json, or .cursor/mcp.json for one project

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

    Cursor completes the sign-in in a browser tab the first time it connects.

  2. Install ElevenLabs and connect it

    Add ElevenLabs to a project in your Toolspoke workspace.

    API key. 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.

    API key
    elevenlabs.io → Profile → API keys. A key carries the scopes it was created with, so a key made read-only will refuse the generate and delete actions here.
  3. Choose what Cursor may do

    All 18 actions start on. Switch off any of the 6 actions that change something in ElevenLabs, and Cursor gets a refusal if it tries.

What you can ask Cursor for

A few of the ElevenLabs actions that only read, so they are safe to try first.

  • Read the account this key belongs to, the user id, the name on it, and a preview of the key itself
  • Fetch one voice by id: its labels, description, sharing terms, verified languages, and saved settings
  • Search the public voice library, voices other people share, which this workspace has not added yet
  • Check a generation started by generate_image

What Cursor can do in ElevenLabs

12 actions read, 4 write and 2 delete or permanently change something.

Reads

  • Get account

    Read the account this key belongs to, the user id, the name on it, and a preview of the key itself. The cheapest proof that the credential works: it takes no argument, spends no credits, and generates nothing. Use get_subscription for the character quota, which this deliberately leaves out.

  • Get subscription

    Read the plan and what is left of it: the tier, characters used against the character limit, when that limit next resets, and how many voice slots are taken. Call this before a long run of generate_speech, the character limit is what stops one part-way through.

  • List models

    List the speech models, with what each one can do, the languages it covers, and its `token_cost_factor`, the multiplier on what a generation costs. Note that generate_speech, generate_image and generate_video take their own separate model ids, listed in their schemas; this endpoint covers the speech models only.

  • List voices

    Search the voices this workspace can already speak with, its own clones, the shared premades, and anything added from the voice library. The `voice_id` of a result is what generate_speech takes. Page with `next_page_token` from the response; narrow with `search` rather than paging through everything.

  • Get voice

    Fetch one voice by id: its labels, description, sharing terms, verified languages, and saved settings. Use it to check what a voice is before spending credits speaking with it.

  • Search shared voices

    Search the public voice library, voices other people share, which this workspace has not added yet. A result cannot be spoken with directly: pass its `public_owner_id` and `voice_id` to add_shared_voice first, and use the `voice_id` that call returns.

  • List history

    List what this workspace has already generated, newest first, the text that was spoken, which voice and model spoke it, when, and how many characters it cost. Use it to find a past generation without paying to make it again.

  • Get history item

    Fetch one history item by id, the full text that was spoken, the voice and model used, the settings, and the character cost. The audio itself is not returned here; it is only downloadable from the ElevenLabs dashboard.

  • Get image generation

    Check a generation started by generate_image. `status` runs pending → generating → completed or failed. A completed response carries `content_url`, a signed link that expires about an hour after it is returned, call this again for a fresh one. A failed generation is not charged.

  • List image generations

    List the image generations this key has made, newest first. Use it to recover the id of something already paid for rather than generating it again, or to find everything still `generating`.

  • Get video generation

    Check a generation started by generate_video. `status` runs pending → generating → completed or failed; video takes minutes rather than seconds, so poll patiently. A completed response carries `content_url`, a signed link that expires about an hour after it is returned, call this again for a fresh one. A failed generation is not charged.

  • List video generations

    List the video generations this key has made, newest first. Use it to recover the id of something already paid for rather than generating it again, or to find everything still `generating`.

Writes

  • Generate speech

    Speak text in a voice and get the audio back in the same call, with the time each character is spoken. Returns `audio_base64` (the MP3 itself, base64-encoded) and `alignment` (`characters` with their `character_start_times_seconds` and `character_end_times_seconds`), which is what captions and lip-sync are timed from. Billed per character against the plan's character limit, so check get_subscription first. Works on every ElevenLabs plan, the free one included. Keep each call to one or two sentences: the gateway reads at most 500,000 characters of any response, and a long clip at a high bitrate would be cut off without warning, so text is capped at 500 characters and the bitrate at 64 kbps. To voice a long script, call this once per line and join the clips.

  • Generate image

    Queue an image generation from a prompt and get back a generation id. This spends credits. Poll get_image_generation with the id until its status is `completed`, then read `content_url`. Each model accepts a different set of options and rejects the rest, so send only what the chosen model takes: `aspect_ratio` is accepted by all of them, `resolution` only by the gemini-3.x and bytedance-seedream models.

  • Generate video

    Queue a video generation from a prompt and get back a generation id. Video is the most expensive thing here, cost rises with duration and resolution, so start short and small. Poll get_video_generation with the id until its status is `completed`, then read `content_url`. Each model accepts a different set of options and rejects the rest: `negative_prompt` and `enhance_prompt` are veo-3.1 only, and the allowed durations, ratios and resolutions differ per model.

  • Add shared voice

    Add a voice from the public library, found with search_shared_voices, to this workspace so generate_speech can use it. It takes one of the workspace's voice slots, get_subscription says how many are left. This adds a voice; it never takes one away.

Deletes or permanently changes

  • Delete voice

    Remove a voice from this workspace for good. A cloned or professional voice cannot be recovered without the original samples, and anything that referenced the voice id stops working. Frees the voice slot it held.

  • Delete history item

    Delete one generation from the workspace's history for good, along with its audio. There is no trash and no undo, and the characters it cost are not refunded.

ElevenLabs in Cursor: questions

Does Cursor see my ElevenLabs credential?

No. The credential is stored encrypted in Toolspoke, and the gateway adds it to each call. Cursor only ever talks to Toolspoke, and the dashboard shows the last four characters of the credential and nothing more.

Can I stop Cursor from changing things in ElevenLabs?

Yes. Each of the 18 actions can be switched off for a project. Turn off the 6 that change something and Cursor can only read.

Will 18 actions crowd Cursor's context?

No. Toolspoke shows Cursor five bridge tools, plus a guide where a toolkit has one, and Cursor searches for the ElevenLabs action it needs when it needs it. Adding more toolkits does not add more tools to its list.

Should it go in every project or just one?

Either. Put the block in ~/.cursor/mcp.json to have Toolspoke in every project, or in .cursor/mcp.json inside one project to keep it there.

What does it cost?

Tool calls are free on every plan, and a ElevenLabs call does not run in a sandbox, so it bills nothing.

Try ElevenLabs in Cursor

Free to start, and tool calls cost nothing on any plan.