Can I use ElevenLabs as a CLI?

Yes. KosmoKrator exposes a ElevenLabs CLI through kosmo integrations:call and the provider shortcut command. Commands can return JSON for scripts, CI, and coding agents.

Can I expose ElevenLabs through MCP?

Yes. The KosmoKrator MCP gateway can expose selected ElevenLabs tools to clients such as Claude Code, Cursor, Codex, and other MCP-compatible tools.

Where are ElevenLabs credentials stored?

ElevenLabs credentials are configured in KosmoKrator's local integration settings and reused by CLI, Lua, and MCP gateway surfaces.

Can I deny write access for ElevenLabs MCP tools?

Yes. Start the gateway with --write=deny for read-only clients, or use ask/allow only for trusted workspaces.

productivity

ElevenLabs Lua API for KosmoKrator Agents

Agent-facing Lua documentation and function reference for the ElevenLabs KosmoKrator integration.

Lua Namespace

Agents call this integration through app.integrations.elevenlabs.*. Use lua_read_doc("integrations.elevenlabs") inside KosmoKrator to discover the same reference at runtime.

Call Lua from the Headless CLI

Use kosmo integrations:lua when a shell script, CI job, cron job, or another coding CLI should run a deterministic ElevenLabs workflow without starting an interactive agent session.

Inline Lua call

kosmo integrations:lua --eval 'dump(app.integrations.elevenlabs.text_speech({voice_id = "example_voice_id", text = "example_text", model_id = "example_model_id", stability = 1, similarity_boost = 1, style = 1, use_speaker_boost = true}))' --json

Read Lua docs headlessly

kosmo integrations:lua --eval 'print(docs.read("elevenlabs"))' --json
kosmo integrations:lua --eval 'print(docs.read("elevenlabs.text_speech"))' --json

Workflow file

Put repeatable logic in a Lua file, then execute it with JSON output for the calling process.

workflow.lua

local elevenlabs = app.integrations.elevenlabs
local result = elevenlabs.text_speech({voice_id = "example_voice_id", text = "example_text", model_id = "example_model_id", stability = 1, similarity_boost = 1, style = 1, use_speaker_boost = true})

dump(result)

Run the workflow

kosmo integrations:lua workflow.lua --json
kosmo integrations:lua workflow.lua --force --json

Namespace note. integrations:lua exposes app.integrations.elevenlabs, app.mcp.*, docs.*, json.*, and regex.*. Use app.integrations.elevenlabs.default.* or app.integrations.elevenlabs.work.* when you configured named credential accounts.

MCP-only Lua

If the script only needs configured MCP servers and does not need ElevenLabs, use the narrower mcp:lua command.

MCP Lua command

# Use mcp:lua for MCP-only scripts; use integrations:lua for this integration namespace.
kosmo mcp:lua --eval 'dump(mcp.servers())' --json

Agent-Facing Lua Docs

This is the rendered version of the full Lua documentation exposed to agents when they inspect the integration namespace.

ElevenLabs Lua API Reference

Namespace: app.integrations.elevenlabs

ElevenLabs tools use the v1 API at https://api.elevenlabs.io/v1 with xi-api-key authentication. Binary audio responses are returned as base64 plus content metadata so agents can save or pass them to downstream tools.

Speech Generation

local audio = app.integrations.elevenlabs.text_to_speech({
  voice_id = "21m00Tcm4TlvDq8ikWAM",
  text = "Hello from ElevenLabs.",
  model_id = "eleven_multilingual_v2",
  stability = 0.5,
  similarity_boost = 0.75
})

local timed = app.integrations.elevenlabs.text_to_speech_with_timestamps({
  voice_id = "21m00Tcm4TlvDq8ikWAM",
  text = "This line needs captions.",
  body = {
    model_id = "eleven_multilingual_v2",
    seed = 1234
  },
  query = {
    output_format = "mp3_44100_128"
  }
})

text_to_speech_with_timestamps returns audio_base64, alignment, and normalized_alignment.

Speech To Speech And Speech To Text

local changed = app.integrations.elevenlabs.speech_to_speech({
  voice_id = "21m00Tcm4TlvDq8ikWAM",
  audio_path = "/tmp/source.wav",
  fields = {
    model_id = "eleven_multilingual_sts_v2",
    remove_background_noise = true
  }
})

local transcript = app.integrations.elevenlabs.speech_to_text({
  audio_path = "/tmp/interview.mp3",
  fields = {
    model_id = "scribe_v1",
    diarize = true,
    keyterms = { "OpenCompany", "KosmoKrator" }
  }
})

Multipart tools require local file paths available to the host.

Sound Effects And Isolation

local sfx = app.integrations.elevenlabs.create_sound_effect({
  text = "Spacious cinematic impact",
  duration_seconds = 2.5,
  prompt_influence = 0.4
})

local clean = app.integrations.elevenlabs.isolate_audio({
  audio_path = "/tmp/noisy.wav",
  fields = {
    file_format = "other"
  }
})

local isolation_history = app.integrations.elevenlabs.list_audio_isolation_history({
  page_size = 50,
  search = "interview"
})

Voices And Models

local voices = app.integrations.elevenlabs.list_voices({})
local voice = app.integrations.elevenlabs.get_voice({ voice_id = "21m00Tcm4TlvDq8ikWAM" })
local settings = app.integrations.elevenlabs.get_voice_settings({ voice_id = "21m00Tcm4TlvDq8ikWAM" })

local updated = app.integrations.elevenlabs.edit_voice_settings({
  voice_id = "21m00Tcm4TlvDq8ikWAM",
  settings = {
    stability = 0.55,
    similarity_boost = 0.8,
    use_speaker_boost = true
  }
})

local models = app.integrations.elevenlabs.get_models({})

create_voice accepts local sample file paths in files. delete_voice permanently removes a voice.

History

local history = app.integrations.elevenlabs.get_history({ page_size = 20 })
local item = app.integrations.elevenlabs.get_history_item({ history_item_id = "hist_123" })
local audio = app.integrations.elevenlabs.get_history_item_audio({ history_item_id = "hist_123" })

delete_history_item({ history_item_id = "..." }) deletes one history item.

Dubbing

local dub = app.integrations.elevenlabs.create_dubbing({
  fields = {
    source_url = "https://example.test/source.mp4",
    target_lang = "es",
    source_lang = "en",
    name = "Spanish trailer dub"
  }
})

local projects = app.integrations.elevenlabs.list_dubbings({})
local project = app.integrations.elevenlabs.get_dubbing({ dubbing_id = dub.dubbing_id })

local transcript = app.integrations.elevenlabs.get_dubbing_transcript({
  dubbing_id = dub.dubbing_id,
  language_code = "es",
  format_type = "json"
})

create_dubbing can also accept files with local paths for file, csv_file, foreground_audio_file, or background_audio_file.

Account And Generic API

local user = app.integrations.elevenlabs.get_current_user({})
local subscription = app.integrations.elevenlabs.get_subscription({})

local raw = app.integrations.elevenlabs.api_get({
  path = "/voices",
  params = { show_legacy = true }
})

Generic helpers are api_get, api_post, and api_delete. Absolute URLs are rejected; pass paths relative to /v1.

Multi-Account Usage

app.integrations.elevenlabs.text_to_speech({ voice_id = "...", text = "Hi" })
app.integrations.elevenlabs.default.text_to_speech({ voice_id = "...", text = "Hi" })
app.integrations.elevenlabs.production.text_to_speech({ voice_id = "...", text = "Hi" })

Raw agent markdown

# ElevenLabs Lua API Reference

Namespace: `app.integrations.elevenlabs`

ElevenLabs tools use the v1 API at `https://api.elevenlabs.io/v1` with `xi-api-key` authentication. Binary audio responses are returned as base64 plus content metadata so agents can save or pass them to downstream tools.

## Speech Generation

```lua
local audio = app.integrations.elevenlabs.text_to_speech({
  voice_id = "21m00Tcm4TlvDq8ikWAM",
  text = "Hello from ElevenLabs.",
  model_id = "eleven_multilingual_v2",
  stability = 0.5,
  similarity_boost = 0.75
})

local timed = app.integrations.elevenlabs.text_to_speech_with_timestamps({
  voice_id = "21m00Tcm4TlvDq8ikWAM",
  text = "This line needs captions.",
  body = {
    model_id = "eleven_multilingual_v2",
    seed = 1234
  },
  query = {
    output_format = "mp3_44100_128"
  }
})
```

`text_to_speech_with_timestamps` returns `audio_base64`, `alignment`, and `normalized_alignment`.

## Speech To Speech And Speech To Text

```lua
local changed = app.integrations.elevenlabs.speech_to_speech({
  voice_id = "21m00Tcm4TlvDq8ikWAM",
  audio_path = "/tmp/source.wav",
  fields = {
    model_id = "eleven_multilingual_sts_v2",
    remove_background_noise = true
  }
})

local transcript = app.integrations.elevenlabs.speech_to_text({
  audio_path = "/tmp/interview.mp3",
  fields = {
    model_id = "scribe_v1",
    diarize = true,
    keyterms = { "OpenCompany", "KosmoKrator" }
  }
})
```

Multipart tools require local file paths available to the host.

## Sound Effects And Isolation

```lua
local sfx = app.integrations.elevenlabs.create_sound_effect({
  text = "Spacious cinematic impact",
  duration_seconds = 2.5,
  prompt_influence = 0.4
})

local clean = app.integrations.elevenlabs.isolate_audio({
  audio_path = "/tmp/noisy.wav",
  fields = {
    file_format = "other"
  }
})

local isolation_history = app.integrations.elevenlabs.list_audio_isolation_history({
  page_size = 50,
  search = "interview"
})
```

## Voices And Models

```lua
local voices = app.integrations.elevenlabs.list_voices({})
local voice = app.integrations.elevenlabs.get_voice({ voice_id = "21m00Tcm4TlvDq8ikWAM" })
local settings = app.integrations.elevenlabs.get_voice_settings({ voice_id = "21m00Tcm4TlvDq8ikWAM" })

local updated = app.integrations.elevenlabs.edit_voice_settings({
  voice_id = "21m00Tcm4TlvDq8ikWAM",
  settings = {
    stability = 0.55,
    similarity_boost = 0.8,
    use_speaker_boost = true
  }
})

local models = app.integrations.elevenlabs.get_models({})
```

`create_voice` accepts local sample file paths in `files`. `delete_voice` permanently removes a voice.

## History

```lua
local history = app.integrations.elevenlabs.get_history({ page_size = 20 })
local item = app.integrations.elevenlabs.get_history_item({ history_item_id = "hist_123" })
local audio = app.integrations.elevenlabs.get_history_item_audio({ history_item_id = "hist_123" })
```

`delete_history_item({ history_item_id = "..." })` deletes one history item.

## Dubbing

```lua
local dub = app.integrations.elevenlabs.create_dubbing({
  fields = {
    source_url = "https://example.test/source.mp4",
    target_lang = "es",
    source_lang = "en",
    name = "Spanish trailer dub"
  }
})

local projects = app.integrations.elevenlabs.list_dubbings({})
local project = app.integrations.elevenlabs.get_dubbing({ dubbing_id = dub.dubbing_id })

local transcript = app.integrations.elevenlabs.get_dubbing_transcript({
  dubbing_id = dub.dubbing_id,
  language_code = "es",
  format_type = "json"
})
```

`create_dubbing` can also accept `files` with local paths for `file`, `csv_file`, `foreground_audio_file`, or `background_audio_file`.

## Account And Generic API

```lua
local user = app.integrations.elevenlabs.get_current_user({})
local subscription = app.integrations.elevenlabs.get_subscription({})

local raw = app.integrations.elevenlabs.api_get({
  path = "/voices",
  params = { show_legacy = true }
})
```

Generic helpers are `api_get`, `api_post`, and `api_delete`. Absolute URLs are rejected; pass paths relative to `/v1`.

## Multi-Account Usage

```lua
app.integrations.elevenlabs.text_to_speech({ voice_id = "...", text = "Hi" })
app.integrations.elevenlabs.default.text_to_speech({ voice_id = "...", text = "Hi" })
app.integrations.elevenlabs.production.text_to_speech({ voice_id = "...", text = "Hi" })
```

Metadata-derived Lua example

local result = app.integrations.elevenlabs.text_speech({voice_id = "example_voice_id", text = "example_text", model_id = "example_model_id", stability = 1, similarity_boost = 1, style = 1, use_speaker_boost = true})
print(result)

Functions

text_speech Write

Convert text to speech audio using an ElevenLabs voice. Returns base64-encoded audio. Choose a voice ID and model ID to control the output.

Lua path: app.integrations.elevenlabs.text_speech
Full name: elevenlabs.elevenlabs_text_to_speech

Parameter	Type	Required	Description
`voice_id`	string	yes	The voice identifier (use list_voices to discover available voices).
`text`	string	yes	The text to convert to speech.
`model_id`	string	no	Model ID to use (e.g., "eleven_multilingual_v2", "eleven_monolingual_v1", "eleven_turbo_v2_5"). Defaults to "eleven_multilingual_v2".
`stability`	number	no	Voice stability (0.0–1.0). Higher values reduce randomness.
`similarity_boost`	number	no	Voice similarity boost (0.0–1.0). Higher values enforce voice similarity.
`style`	number	no	Style exaggeration (0.0–1.0). Higher values increase expressiveness.
`use_speaker_boost`	boolean	no	Enable speaker boost for enhanced clarity.

text_speech_with_timestamps Write

Convert text to speech with character timing.

Lua path: app.integrations.elevenlabs.text_speech_with_timestamps
Full name: elevenlabs.elevenlabs_text_to_speech_with_timestamps

Parameter	Type	Required	Description
No parameters.

speech_speech Write

Transform an audio file into another voice.

Lua path: app.integrations.elevenlabs.speech_speech
Full name: elevenlabs.elevenlabs_speech_to_speech

Parameter	Type	Required	Description
No parameters.

speech_text Read

Transcribe audio or video.

Lua path: app.integrations.elevenlabs.speech_text
Full name: elevenlabs.elevenlabs_speech_to_text

Parameter	Type	Required	Description
No parameters.

create_sound_effect Write

Generate a sound effect from text.

Lua path: app.integrations.elevenlabs.create_sound_effect
Full name: elevenlabs.elevenlabs_create_sound_effect

Parameter	Type	Required	Description
No parameters.

isolate_audio Write

Remove background noise from audio.

Lua path: app.integrations.elevenlabs.isolate_audio
Full name: elevenlabs.elevenlabs_isolate_audio

Parameter	Type	Required	Description
No parameters.

list_audio_isolation_history Read

List audio isolation generations.

Lua path: app.integrations.elevenlabs.list_audio_isolation_history
Full name: elevenlabs.elevenlabs_list_audio_isolation_history

Parameter	Type	Required	Description
No parameters.

list_voices Read

List all available voices in your ElevenLabs account, including pre-made and cloned voices. Use this to discover voice IDs for text-to-speech.

Lua path: app.integrations.elevenlabs.list_voices
Full name: elevenlabs.elevenlabs_list_voices

Parameter	Type	Required	Description
No parameters.

get_voice Read

Get detailed information about a specific ElevenLabs voice, including its settings, labels, and fine-tuning info.

Lua path: app.integrations.elevenlabs.get_voice
Full name: elevenlabs.elevenlabs_get_voice

Parameter	Type	Required	Description
`voice_id`	string	yes	The voice identifier to look up.

get_voice_settings Read

Get voice settings.

Lua path: app.integrations.elevenlabs.get_voice_settings
Full name: elevenlabs.elevenlabs_get_voice_settings

Parameter	Type	Required	Description
No parameters.

edit_voice_settings Write

Edit voice settings.

Lua path: app.integrations.elevenlabs.edit_voice_settings
Full name: elevenlabs.elevenlabs_edit_voice_settings

Parameter	Type	Required	Description
No parameters.

create_voice Write

Create a new voice clone in ElevenLabs. Provide a name, optional audio sample file paths or base64-encoded data, and an optional description.

Lua path: app.integrations.elevenlabs.create_voice
Full name: elevenlabs.elevenlabs_create_voice

Parameter	Type	Required	Description
`name`	string	yes	Name for the new voice.
`files`	array	no	Array of audio sample file paths or base64-encoded audio strings for voice cloning.
`description`	string	no	Optional description of the voice (e.g., "Warm female narrator").

delete_voice Write

Permanently delete a voice from your ElevenLabs account. This action cannot be undone.

Lua path: app.integrations.elevenlabs.delete_voice
Full name: elevenlabs.elevenlabs_delete_voice

Parameter	Type	Required	Description
`voice_id`	string	yes	The voice identifier to delete.

get_models Read

List all available ElevenLabs text-to-speech models, including their IDs, names, and language support. Use model IDs when calling text_to_speech.

Lua path: app.integrations.elevenlabs.get_models
Full name: elevenlabs.elevenlabs_get_models

Parameter	Type	Required	Description
No parameters.

get_history Read

Browse your ElevenLabs generation history. Returns a paginated list of past text-to-speech requests with metadata.

Lua path: app.integrations.elevenlabs.get_history
Full name: elevenlabs.elevenlabs_get_history

Parameter	Type	Required	Description
`page_size`	integer	no	Number of history items per page (default: 20, max: 100).
`start_after`	integer	no	History item ID to start after (for cursor-based pagination).

get_history_item Read

Get one history item.

Lua path: app.integrations.elevenlabs.get_history_item
Full name: elevenlabs.elevenlabs_get_history_item

Parameter	Type	Required	Description
No parameters.

get_history_item_audio Read

Download history item audio.

Lua path: app.integrations.elevenlabs.get_history_item_audio
Full name: elevenlabs.elevenlabs_get_history_item_audio

Parameter	Type	Required	Description
No parameters.

delete_history_item Write

Delete one history item.

Lua path: app.integrations.elevenlabs.delete_history_item
Full name: elevenlabs.elevenlabs_delete_history_item

Parameter	Type	Required	Description
No parameters.

create_dubbing Write

Create a dubbing project.

Lua path: app.integrations.elevenlabs.create_dubbing
Full name: elevenlabs.elevenlabs_create_dubbing

Parameter	Type	Required	Description
No parameters.

list_dubbings Read

List dubbing projects.

Lua path: app.integrations.elevenlabs.list_dubbings
Full name: elevenlabs.elevenlabs_list_dubbings

Parameter	Type	Required	Description
No parameters.

get_dubbing Read

Get one dubbing project.

Lua path: app.integrations.elevenlabs.get_dubbing
Full name: elevenlabs.elevenlabs_get_dubbing

Parameter	Type	Required	Description
No parameters.

delete_dubbing Write

Delete a dubbing project.

Lua path: app.integrations.elevenlabs.delete_dubbing
Full name: elevenlabs.elevenlabs_delete_dubbing

Parameter	Type	Required	Description
No parameters.

get_dubbing_transcript Read

Get a dubbing transcript.

Lua path: app.integrations.elevenlabs.get_dubbing_transcript
Full name: elevenlabs.elevenlabs_get_dubbing_transcript

Parameter	Type	Required	Description
No parameters.

get_current_user Read

Get your ElevenLabs account details, including subscription tier, character usage, and remaining quota.

Lua path: app.integrations.elevenlabs.get_current_user
Full name: elevenlabs.elevenlabs_get_current_user

Parameter	Type	Required	Description
No parameters.

get_subscription Read

Get subscription and quota details.

Lua path: app.integrations.elevenlabs.get_subscription
Full name: elevenlabs.elevenlabs_get_subscription

Parameter	Type	Required	Description
No parameters.

api_get Read

Call an ElevenLabs GET endpoint.

Lua path: app.integrations.elevenlabs.api_get
Full name: elevenlabs.elevenlabs_api_get

Parameter	Type	Required	Description
No parameters.

api_post Write

Call an ElevenLabs POST endpoint.

Lua path: app.integrations.elevenlabs.api_post
Full name: elevenlabs.elevenlabs_api_post

Parameter	Type	Required	Description
No parameters.

api_delete Write

Call an ElevenLabs DELETE endpoint.

Lua path: app.integrations.elevenlabs.api_delete
Full name: elevenlabs.elevenlabs_api_delete

Parameter	Type	Required	Description
No parameters.