Integrations · Playbook

What an AI coworker
can do in Elevenlabs.

The Elevenlabs integration exposes 24 actions — 23 read,1 write — that any agent, team or AI employee can use once Elevenlabs is connected at the organization level. Every write can be gated behindhuman approval, and every call lands in theaudit trail.

Read actions (23)

Safe to run freely — they observe, fetch and search.

ActionFunctionKey params
Add knowledge base to agentadd_knowledge_base_to_agentagent_id, knowledge_base_name
Check subscriptioncheck_subscription
Compose musiccompose_music
Create agentcreate_agentname, first_message, system_prompt
Create composition plancreate_composition_planprompt
Create voice from previewcreate_voice_from_previewgenerated_voice_id, voice_name, voice_description
Get agentget_agentagent_id
Get conversationget_conversationconversation_id
Get voiceget_voicevoice_id
Isolate audioisolate_audioinput_file_path
List agentslist_agents
List conversationslist_conversations
List modelslist_models
List phone numberslist_phone_numbers
Make outbound callmake_outbound_callagent_id, agent_phone_number_id, to_number
Play audioplay_audioinput_file_path
Search voice librarysearch_voice_library
Search voicessearch_voices
Speech to speechspeech_to_speechinput_file_path
Speech to textspeech_to_textinput_file_path
Text to sound effectstext_to_sound_effectstext
Text to speechtext_to_speechtext
Text to voicetext_to_voicevoice_description

Write actions (1)

They change things. Gate the sensitive ones with approval policies; approvals are single-use.

ActionFunctionKey params
Voice clonevoice_clonename, files

Put Elevenlabs in an AI coworker's hands.

Browse templates and solution packs that work with Elevenlabs, or start from the problem you're trying to fix.

Browse templates