Vapi
for LuumenAI
Read call history and assistant config, then change voice agents with approval
Connect Vapi and Luumen can read your voice agents the way you would: list assistants, pull a specific assistant's transcriber, voice, and model settings, page through recent calls, and run analytics queries. Changes such as updating an assistant or creating an eval are previewed first and need your approval by default. Luumen authenticates with an API key.
The Vapi toolbox
38 tools: 19 read, 19 write. Reads answer instantly. Writes require approval by default. Everything is logged.
- ReadList CallsList calls from Vapi.
- ReadGet ChatFetch chat details by ID.
- ReadCreate Analytics QueriesCreate and execute analytics queries on VAPI data.
- ReadGet EvalRetrieve an eval by its ID.
- ReadGet AssistantRetrieve a specific assistant by ID from Vapi.
- ReadGet CallFetch call details by ID.
- ReadGet FileRetrieve a file by its ID from Vapi.
- ReadList Monitoring PoliciesRetrieve monitoring policies from Vapi.
- ReadList Structured OutputsList structured outputs with optional filtering.
- ReadGet InsightsRetrieve insights from VAPI.
- ReadList AssistantsList all assistants in your VAPI organization.
- ReadList ChatsRetrieve a list of chat conversations from VAPI.
- ReadList EvalsRetrieve a paginated list of evals from Vapi.
- ReadList Phone NumbersList phone numbers in the Vapi organization.
- ReadList Provider ResourcesRetrieve provider resources from Vapi (e.g., 11labs pronunciation dictionaries).
- ReadList ScorecardsRetrieve a paginated list of scorecards from Vapi.
- ReadList SessionsRetrieve a paginated list of sessions from VAPI.
- ReadGet ToolFetch tool details by ID.
- ReadTest Code Tool ExecutionTest TypeScript code execution in Vapi's code tool environment.
- WriteUpdate AssistantUpdate an existing Vapi assistant configuration. Approval by default
- WriteDelete ChatDelete a chat by its ID from Vapi. Approval by default
- WriteCreate AssistantCreate a new Vapi assistant with specified transcriber, voice, and AI model configurations. Approval by default
- WriteCreate EvalCreate an eval for testing conversation flows. Approval by default
- WriteCreate OpenAI ChatCreate an OpenAI-compatible chat using the Vapi API. Approval by default
- WriteCreate Monitoring PolicyCreate a monitoring policy in VAPI. Approval by default
- WriteCreate Provider ResourceCreate an 11Labs pronunciation dictionary resource. Approval by default
- WriteCreate ScorecardCreate a scorecard for observability and evaluation. Approval by default
- WriteDelete CallDelete a call by its unique identifier. Approval by default
- WriteDelete EvalDelete an eval by ID. Approval by default
- WriteDelete Phone NumberDelete a phone number from Vapi. Approval by default
- WriteDelete Eval RunDelete an eval run by its ID from Vapi. Approval by default
- WriteUpdate EvalUpdate an existing eval in Vapi. Approval by default
- WriteUpdate InsightUpdate an existing insight configuration in VAPI. Approval by default
- WriteCreate Phone NumberCreate a phone number with VAPI. Approval by default
- WriteCreate SessionCreate a new session in Vapi. Approval by default
- WriteUpdate ToolUpdate an existing Vapi tool configuration. Approval by default
- WriteUpdate Phone NumberUpdate an existing phone number configuration in VAPI. Approval by default
- WriteUpload FileUpload a file to Vapi Knowledge Base. Approval by default
One prompt, start to finish
What a governed Vapi run looks like inside Luumen.
Questions
Authorize once with API token. Luumen lists the scopes each action needs before you approve the connection, and credentials never appear in the chat.
Read actions answer immediately. Anything that writes — update assistant, delete chat, create assistant, create eval, and more — is shown as a plan and requires approval by default, including the 5 actions classified as destructive. Administrators configure that per tool, so you decide exactly which actions can ever run unattended.
You decide. Actions are granted per agent, skill, and team, and per environment — production is not staging. Read access can be broad while writes stay narrow.
Every call to Vapi — read or write, approved or declined — is recorded with the actor, the input, and the result, and can be linked to the ticket or change record.
Often connected alongside
All integrations
Adobe Marketing Agent MCPAdobe Experience Platform context, sandbox switches, and async task status
AffindaRead parsed documents, collections, and annotations — change them with approval
Agent MailGive the agent its own mailbox for reading threads and sending updates
AI/ML APIInspect assistants, threads, and runs — and change them as approved stepsPut Vapi to work with Luumen
Connect in minutes. Every action scoped, approved, and audited from day one.