Integration

Beyond Presence integration

Beyond Presence runs real-time conversational video agents on lifelike AI avatars. Use it to list stock and custom avatars, create, update and delete managed agents (prompt, greeting, language, LLM, voice, knowledge files), read the conversations those agents held with their transcripts, manage text knowledge files, list external LLM or ElevenLabs configurations, and read plan usage and speech-to-video session history. Users talk to an agent in its managed room or Studio embed, not through these tools. Not exposed: LiveKit room creation (it returns join credentials), starting speech-to-video sessions (they need your own LiveKit or Daily room token), creating external API configurations (they take third-party API keys), chunked PDF uploads, and the API key check; LLM endpoint and media server URLs and conversation tags are not returned. Concurrent sessions and included minutes depend on the plan (Free: 1 session, 40 minutes); a 402 means the usage allowance is used up, a 429 the concurrency limit. Agent conversations are billed per minute (1 credit = 1 minute) against the plan's included monthly minutes (Free 40, Starter 280, Growth 1,490, Scale 4,000), then at the plan's overage rate (EUR 0.0875 to 0.175 per minute) when usage-based billing is on; Free plan conversations stop after 3 minutes. No tool here starts a conversation or session; max_session_length_minutes caps how long each conversation lasts. Authentication is an API key from the Beyond Presence Studio, sent in the x-api-key header.

Connect Beyond Presence How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Beyond Presence.Built from the 21 Beyond Presence tools below, on the connection you add once.

What you can do with Beyond Presence

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Agent, 5 operations

create_agentCreate agentPOSTCreate a managed video agent from an avatar (see list_avatars) and a system prompt. Users can then talk to it in its managed room, using plan minutes. Returns the agent.
delete_agentDelete agentDELETEDelete an agent for real and permanently; it can no longer hold conversations. Conversations it already held are not affected.
get_agentGet agentGETGet one agent by id (from list_agents or create_agent) with its prompt, greeting, language, LLM, voice, capabilities and knowledge files.
list_agentsList agentsGETList your managed agents, most recent first, with their prompt, avatar, language, LLM and voice settings. Cursor-paginated: pass next_cursor as cursor for the next page.
update_agentUpdate agentPATCHChange an agent's settings. Only the fields you send change; every other field keeps its current value. Answers with no content on success; read the result with get_agent.

Knowledge file, 4 operations

create_knowledge_fileCreate knowledge filePOSTCreate a text knowledge file for agents to answer from. It comes back as processing; poll get_knowledge_file until status is available, then attach its id to an agent with update_agent.
delete_knowledge_fileDelete knowledge fileDELETEDelete a knowledge file for real. Fails while any agent still references it: remove it from that agent's knowledge_file_ids with update_agent first.
get_knowledge_fileGet knowledge fileGETGet one knowledge file by id with its status and extracted text (shortened for a large file, see is_text_complete). Poll it after create_knowledge_file until status is available or failed.
list_knowledge_filesList knowledge filesGETList your knowledge files with their status (to-upload, processing, available, failed) and extracted text. Cursor-paginated: pass next_cursor as cursor for the next page.

External api, 3 operations

delete_external_apiDelete external apiDELETEDelete an external API configuration and its stored secret, for real. Fails while any agent still uses it: change that agent's llm with update_agent first.
get_external_apiGet external apiGETGet one external API configuration by id (from list_external_apis): its kind and name. Its secret and endpoint URL are not returned.
list_external_apisList external apisGETList your external API configurations (your own OpenAI-compatible LLM or ElevenLabs account) by id, kind and name; their secrets and endpoint URLs are not returned. Use an id as llm.api_id on an agent.

Avatar, 2 operations

get_avatarGet avatarGETGet one avatar by id (from list_avatars): its name, visibility (public stock or private custom) and status.
list_avatarsList avatarsGETList the avatars you can give an agent: public stock avatars and your own custom ones, with their training status. Cursor-paginated: pass next_cursor as cursor for the next page.

Conversation, 2 operations

get_conversationGet conversationGETGet one conversation by id (from list_conversations): its agent, user name and email if given, and status with start and end times. Read the transcript with list_conversation_messages.
list_conversationsList conversationsGETList the conversations your agents held, across all agents, most recent first, with the user's name and email if given and the status. Pass next_cursor as cursor for the next page.

Session, 2 operations

get_sessionGet sessionGETGet one speech-to-video session by id (from list_sessions): its transport, avatar and status with start and end times.
list_sessionsList sessionsGETList your speech-to-video sessions (an avatar joined to your own LiveKit or Daily room), most recent first, with avatar and status. Pass next_cursor as cursor for the next page.

Usage, 2 operations

get_usageGet usageGETGet your organization's usage in the current billing period, in seconds: included, used, remaining and overage. Needs an API key of the default workspace; usage can take a few minutes to appear.
list_usage_sessionsList usage sessionsGETList the billed seconds of each ended speech-to-video session created between start and end (UTC days, at most 366 days and 10,000 sessions). Needs a default-workspace API key.

Message, 1 operation

list_conversation_messagesList conversation messagesGETGet the transcript of a conversation in chronological order, each message attributed to the agent (ai) or the user. Takes a conversation id from list_conversations.

Use Beyond Presence from Claude Code, Cursor or Codex

Connect Beyond Presence in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Beyond Presence operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Beyond Presence

Create your workspace
Sign up free. Nothing to install.
Connect Beyond Presence once
Add your Beyond Presence connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.