ElevenLabs MCP integration for AI agents.
Connect AI agents to ElevenLabs through 14 structured actions, including get audio from history item, create sound effect, delete history item, and text to s…
- ACTION CONTRACTS
- 14
- AUTHENTICATION
- API key
- SOURCE STATUS
- Provider-backed
CATALOG SOURCE REVIEWED AUGUST 23, 2026 / ACTION NAMES AND SCHEMAS DERIVED FROM WORKING MACHINES PROVIDER SOURCE
CAPABILITY PROFILE
What the ElevenLabs integration exposes
The ElevenLabs Agent App exposes 14 provider-backed actions for ai and design & media work. Its current contract lets an authorized agent download the audio for one elevenlabs history item and upload the binary result to connector transit storage, generate a sound effect from a text prompt and upload the binary audio result to connector transit storage, and delete one elevenlabs history item by history item id. These operations are called through Working Machines as typed capabilities rather than through browser navigation or copied UI steps.
A connection uses API key. Before execution, the agent can inspect the selected action, its required fields, declared scopes, and expected output contract. Provider credentials remain inside the Working Machines runtime; the calling agent receives the capability and its structured result, not the underlying secret.
For reliable operation, start with the narrowest action that satisfies the task, resolve stable provider identifiers before changing state, and validate the returned object or status after execution. Availability still depends on the connected ElevenLabs account, granted provider permissions, workspace policy, region, plan, and upstream API behavior.
VERIFIED ACTION SAMPLE
Real ElevenLabs capabilities.
Showing 12 of 14 actions. Risk labels are conservative signals based on operation names, not substitutes for provider documentation or runtime policy.
get_audio_from_history_itemREADDownload the audio for one ElevenLabs history item and upload the binary result to connector transit storage.
INPUTS: historyItemId
create_sound_effectWRITEGenerate a sound effect from a text prompt and upload the binary audio result to connector transit storage.
INPUTS: text / outputFormat / loop / durationSeconds / promptInfluence
delete_history_itemHIGH IMPACTDelete one ElevenLabs history item by history item ID.
INPUTS: historyItemId
text_to_speechREVIEWGenerate speech audio from text by calling ElevenLabs text-to-speech and uploading the binary result to connector transit storage.
INPUTS: voiceId / text / modelId / outputFormat / optimizeStreamingLatency
get_generated_itemsREADList generated ElevenLabs history items with pagination and optional voice filtering.
INPUTS: pageSize / voiceId / startAfterHistoryItemId
get_history_item_by_idREADGet one ElevenLabs history item by history item ID without downloading its audio.
INPUTS: historyItemId
get_modelsREADList the available ElevenLabs models and their text-to-speech capabilities.
get_user_infoREADGet the current ElevenLabs user profile together with the embedded subscription snapshot.
get_user_subscription_infoREADGet the current ElevenLabs subscription details for the authenticated user.
get_voiceREADGet one ElevenLabs voice by voice ID, with optional settings included.
INPUTS: voiceId / withSettings
get_voice_settingsREADGet the synthesis settings configured for one ElevenLabs voice.
INPUTS: voiceId
get_voicesREADList the available ElevenLabs voices with their key metadata and settings.
INPUT CONTRACTS
Know what the action needs before it runs.
get_audio_from_history_item
historyItemIdOPTIONAL- The history item ID whose audio should be downloaded.
create_sound_effect
textREQUIRED- The prompt text that will be converted into a sound effect.
outputFormatOPTIONAL- The output audio format for generated sound effects.
loopOPTIONAL- Whether to create a smoothly looping sound effect.
durationSecondsOPTIONAL- The generated sound duration in seconds.
promptInfluenceOPTIONAL- How strongly the generation should follow the prompt.
delete_history_item
historyItemIdOPTIONAL- The history item ID to delete.
text_to_speech
voiceIdREQUIRED- The voice ID to use for synthesis.
textREQUIRED- The input text to synthesize.
modelIdOPTIONAL- The model ID to use for synthesis.
outputFormatOPTIONAL- The output audio format. ElevenLabs formats are named as codec_sample_rate_bitrate, such as mp3_44100_128.
optimizeStreamingLatencyOPTIONAL- The streaming latency optimization level.
PROVIDER-SPECIFIC WORKFLOWS
Jobs this Agent App can support
Inspect Get Audio From History Item
Download the audio for one ElevenLabs history item and upload the binary result to connector transit storage. Use this as a bounded discovery step, retain the returned identifier, and avoid expanding the read beyond the task's stated scope.
get_audio_from_history_itemControl Create Sound Effect
Generate a sound effect from a text prompt and upload the binary audio result to connector transit storage. Resolve the target first, present material changes for confirmation, and make retries idempotent where the provider supports it.
create_sound_effectVerify with Get Generated Items
List generated ElevenLabs history items with pagination and optional voice filtering. Compare the returned provider state with the intended outcome and preserve stable IDs or canonical links in the run record.
get_generated_itemsSAFETY BOUNDARY
Operate ElevenLabs with explicit limits
- Authorize ElevenLabs with API key and grant only the provider access required by the selected actions.
- 2 actions are change-capable by name. Confirm the target identity and material parameters before allowing a write.
- Treat `delete_history_item` as high-impact operations and require an explicit approval boundary.
- No provider scope string is declared on the sampled actions. Verify the connected account's actual permissions in ElevenLabs rather than assuming unrestricted access.
- After a call, inspect the structured result and execution record before reporting that the ElevenLabs task completed successfully.
CONNECTION MODEL
API key
ElevenLabs API key sent with the xi-api-key header. Create or manage keys from Developers > API Keys.
Working Machines stores provider credentials behind the execution boundary. An agent can use an authorized connection identity, but catalog discovery alone does not reveal OAuth tokens, API keys, or provider secrets.
- READ SIGNALS
- 10
- WRITE SIGNALS
- 1
- HIGH IMPACT
- 1
- REVIEW SIGNALS
- 2
EVIDENCE AND AVAILABILITY
Provider reference
Action names, input fields, authentication types, and counts on this page are generated from the Working Machines provider catalog. Provider behavior, quotas, object semantics, account eligibility, and regional availability remain governed by ElevenLabs.