Integration

Famulor integration

Famulor runs AI voice and messaging agents for a business. Build and update phone and chat assistants (prompt, voice, languages, tools, calendars, knowledge), place outbound calls, run calling campaigns over leads and segments, send SMS, reply in email and messaging conversations, and read call history with transcripts, summaries and recordings. Manage knowledge bases (documents, website crawls, FAQs and search), reusable tools, native bookings and event types, contacts and their memory, the opt-out (suppression) list, scheduled callbacks, Milian Missions, dashboards, widgets and signed webhook subscriptions. Calls, SMS, campaign dialing, crawls, mission runs and Milian questions spend workspace credits. Not exposed: white-label and reseller administration, API keys and team or workspace administration, account security settings, billing payments and subscription changes, phone number purchase and regulatory verification, SIP trunks and carrier connections, the Famulor Loop phone system, OAuth and channel connection flows (WhatsApp, Messenger, Slack, Calendly, Acuity), Shopify, avatars and voice cloning, automations (their graph and answers are undocumented), QA runs, assistant versions and simulations, live call control, and audio or video downloads. White-label tenant domains are not supported: calls go to app.famulor.io. Authentication is a workspace API key (fam_...) created in Famulor under Settings, API & MCP and sent as a Bearer token; REST access needs API Access on the workspace plan, and a key may be limited to scopes.

Connect Famulor How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Famulor.Built from the 130 Famulor tools below, on the connection you add once.

What you can do with Famulor

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.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

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

Assistant, 6 operations

create_assistantCreate assistantPOSTAgents: ask firstCreates an assistant. Provide `name`, or pass a visible catalog `template_id` to resolve its prompt, greeting, language, variables and optional Flow JSON server-side. Explicit request fields override the template seed.
delete_assistantDelete assistantDELETEAgents: ask firstDeletes the assistant permanently. Phone numbers pointing at it are unassigned. Not possible while one of its calls is live: wait until the call has ended, then retry. Open translation rooms that nobody has joined are ended.
get_assistantGet assistantGETAgents: ask firstReturns one assistant's full configuration: prompt, greeting, voice, speech and language settings, timeouts, knowledge base, built-in tools, memory, consent and analysis settings. Required scope: assistants:read.
list_assistantsList assistantsGETAgents: ask firstPaginated list of assistants in your workspace, newest first. Required scope: `assistants:read` (keys without scope restrictions have full access).
list_knowledge_base_assistantsList knowledge base assistantsGETAgents: ask firstLists the assistants that search this knowledge base during calls and chats, sorted by name. Each assistant uses at most one knowledge base. Required scope: `knowledge:read` or `assistants:read` (keys without scope restrictions have full access).
update_assistantUpdate assistantPATCHAgents: ask firstUpdates an assistant: only the fields you send change (leave a field out to keep its value). Every update creates a version snapshot. Required scope: assistants:write.

Campaign, 6 operations

create_campaignCreate campaignPOSTAgents: ask firstCreates a campaign in status `draft`. Dialer settings (concurrency, retries, calling windows, AMD/voicemail behavior) can be set right away. The plan's campaign limit is enforced (archived campaigns do not count).
delete_campaignDelete campaignDELETEAgents: ask firstDeletes the campaign. Its leads are kept (their campaign reference is cleared). Required scope: `campaigns:write` (keys without scope restrictions have full access).
get_campaignGet campaignGETAgents: ask firstCampaign details including `total_leads`. Required scope: `campaigns:read` (keys without scope restrictions have full access).
list_campaignsList campaignsGETAgents: ask firstLists the workspace's calling campaigns (paginated with limit and offset): status, assistant, channel, dialer settings, calling windows and timezone. Required scope: campaigns:read.
start_campaignStart campaignPOSTAgents: ask firstStarts a campaign: sets it to running and dials the first wave of leads at once (real calls, billed), then further waves on schedule within calling windows. Required scope: campaigns:write.
stop_campaignStop campaignPOSTAgents: ask firstPauses a running campaign (status `paused`). Calls already in progress finish normally; the dialer stops dialing new leads. Returns 409 if the campaign is not running.

Lead, 6 operations

create_audience_contactCreate audience contactPOSTAgents: ask firstCreate an unassigned Audience contact, or merge user-defined tags and missing identity fields into an existing contact with the same phone or email. Required scope: `leads:write` (legacy `calls:write` is also accepted).
get_campaign_leadGet campaign leadGETAgents: ask firstReturns one lead's editable contact data (personal data) and user-managed `custom_fields` plus a separate read-only `system_variables` object. Required scope: `leads:read`.
list_audience_contactsList audience contactsGETAgents: ask firstLists contacts across all campaigns (personal data) with channel and tenant-scoped Call QA filters. Channel matches use linked calls, SMS, email, messaging conversations and channel profiles; repeating `channel` matches any selected channel.
list_campaign_leadsList campaign leadsGETAgents: ask firstLeads of a campaign (personal data: their contact details), newest first. Every lead includes the derived `display_status`. Required scope: `leads:read` (keys without scope restrictions have full access).
list_segment_leadsList segment leadsGETAgents: ask firstLeads matching the segment's saved filters (personal data), newest first, paginated, each with the derived `display_status`. Resolution is capped at 5000 leads; `meta.pagination.total` reflects that cap, not an unbounded count.
remove_campaign_leadRemove campaign leadDELETEAgents: ask firstRemoves the lead from campaign execution while retaining the Audience contact, calls, and completed campaign history. This operation never permanently deletes the contact.

Tool, 6 operations

archive_toolArchive toolDELETEAgents: ask firstArchive an unused tool. Active assignments return `409 tool_in_use`; disabling remains available at any time. Pass the current revision as `expected_revision` or `If-Match`. Required scope: `assistants:write`.
create_toolCreate toolPOSTAgents: ask firstCreate one centrally managed API, MCP, or built-in tool. Requires an admin-level credential because configs may contain secrets. Assign it via `set_assistant_tools`. Required scope: `assistants:write`.
get_toolGet toolGETAgents: ask firstRead a single reusable tool. Secret values in `config` are always masked as `•••`. Required scope: `assistants:read` (keys without scope restrictions have full access).
list_assistant_toolsList assistant toolsGETAgents: ask firstList assigned central tools. `meta.revision` is required for an atomic replacement. Required scope: `assistants:read`.
list_toolsList toolsGETAgents: ask firstList the central API, MCP, and built-in Tool Registry. Each item includes its optimistic `revision`, `usage_count`, OAuth connection health, and latest execution result; secret values are masked. Required scope: `assistants:read`.
set_assistant_toolsSet assistant toolsPUTAgents: ask firstAtomically replace the assistant's complete tool set. `expected_revision` prevents overwriting concurrent assignment changes. Required scope: `assistants:write`.

Booking, 5 operations

cancel_bookingCancel bookingPOSTAgents: ask firstCancel a confirmed booking. Sends a `METHOD:CANCEL` ICS email to the invitee and removes the pushed external calendar event (both best-effort). Idempotent for already-cancelled bookings; bookings that already started return `400`.
create_bookingCreate bookingPOSTAgents: ask firstBooks a free start time of an event type for an invitee (take start from list_booking_slots). Sends the invitee a confirmation email and pushes the event to connected calendars. Required scope: bookings:write.
get_bookingGet bookingGETAgents: ask firstRead a single booking. Required scope: `bookings:read` (keys without scope restrictions have full access).
list_bookingsList bookingsGETAgents: ask firstLists bookings of Famulor's native booking engine (paginated with limit and offset): event type, invitee name, email and phone, start and end time, status and answers to booking questions. Personal data. Required scope: bookings:read.
reschedule_bookingReschedule bookingPOSTAgents: ask firstMoves a confirmed booking to a new start time (validated against the event type's availability) and notifies the invitee by email. Required scope: bookings:write.

Crawl source, 5 operations

create_crawl_sourceCreate crawl sourcePOSTAgents: ask firstRegister a website as a crawl source for a knowledge base. You can provide an optional custom `name`; when omitted, the source receives a URL-based display name that can be edited later.
delete_crawl_sourceDelete crawl sourceDELETEAgents: ask firstDelete a website crawl source. Documents already created by past crawls are kept; only future syncs stop. Required scope: `knowledge:write` (keys without scope restrictions have full access).
get_crawl_sourceGet crawl sourceGETAgents: ask firstFetch a single website crawl source, including its custom display name, last run status, `pages_crawled`, `credits_spent` and `last_error`. Required scope: `knowledge:read` (keys without scope restrictions have full access).
list_crawl_sourcesList crawl sourcesGETAgents: ask firstList the website crawl sources of a knowledge base, including each source's editable custom display name. Each source crawls pages under a root URL (same host only, static HTML, no JavaScript rendering) into knowledge documents.
run_crawl_sourceRun crawl sourcePOSTAgents: ask firstCrawl a website source immediately (synchronous). Fetches pages under the root URL, re-indexes changed pages (unchanged pages are skipped via content hash, no re-embedding, no credits) and charges credits per crawled page.

Booking event type, 4 operations

create_booking_event_typeCreate booking event typePOSTAgents: ask firstCreate an event type. The slug is unique within the workspace and becomes the public booking page URL `/book/{workspace}/{slug}`. Requires an admin-level credential; requires Calendar integrations in the workspace plan.
delete_booking_event_typeDelete booking event typeDELETEAgents: ask firstPermanently delete an event type. All its bookings are deleted with it (ON DELETE CASCADE). Required scope: `bookings:write` (keys without scope restrictions have full access).
get_booking_event_typeGet booking event typeGETAgents: ask firstRead a single event type. Required scope: `bookings:read` (keys without scope restrictions have full access).
list_booking_event_typesList booking event typesGETAgents: ask firstList the event types of the native booking engine. Each active event type has a public, iframe-embeddable booking page at `/book/{workspace}/{slug}` and can be attached to assistants via a `native` calendar integration.

Integration, 4 operations

get_integrationGet integrationGETAgents: ask firstRead a single calendar integration. Secret values are masked. Required scope: `integrations:read` (keys without scope restrictions have full access).
list_assistant_integrationsList assistant integrationsGETAgents: ask firstList the calendar integrations assigned to an assistant. Each assignment adds `check_availability` and `book_appointment` (Calendly link mode: `send_booking_link`) plus provider-supported appointment lookup, cancellation, and rescheduling tools.
list_integrationsList integrationsGETAgents: ask firstList all calendar integrations of the account (Cal.com, Calendly, Acuity Scheduling, Google Calendar, Outlook, native booking engine). Secret values in configs are always masked.
set_assistant_integrationsSet assistant integrationsPUTAgents: ask firstREPLACE the integration assignments of an assistant with the given set. Pass an empty array to remove all assignments (possible even after a plan downgrade); assigning requires a plan that includes calendar integrations.

Knowledge base, 4 operations

create_knowledge_baseCreate knowledge basePOSTAgents: ask firstCreates an empty knowledge base. Attach it to an assistant via the assistant's `knowledgebase_id`. Required scope: `knowledge:write` (keys without scope restrictions have full access).
delete_knowledge_baseDelete knowledge baseDELETEAgents: ask firstDeletes the knowledge base including all its documents and chunks. Required scope: `knowledge:write` (keys without scope restrictions have full access).
get_knowledge_baseGet knowledge baseGETAgents: ask firstReturns one knowledge base by id. Required scope: knowledge:read.
list_knowledge_basesList knowledge basesGETAgents: ask firstLists the workspace's knowledge bases (paginated with limit and offset); attach one to an assistant through its knowledgebase_id. Required scope: knowledge:read.

Memory, 4 operations

erase_contact_memoryErase contact memoryDELETEAgents: ask firstPermanently deletes all stored summaries, preferences, and remembered variables for the contact. A content-free deletion audit remains; the contact and exact channel identities remain. Required scope: `leads:write` or `calls:write`.
get_contact_memoryGet contact memoryGETAgents: ask firstReturns the contact, shared and assistant-specific memories, exact channel identities and recent audit events (personal data). Required scope: `leads:read` or `calls:read`.
get_memory_settingsGet memory settingsGETAgents: ask firstReads the workspace defaults for caller memory. Assistants with `memory_mode: "inherit"` (see `update_assistant`) follow `memory_enabled_default`. Required scope: `settings:read`.
list_contact_memoriesList contact memoriesGETAgents: ask firstLists one consent- and retention-aware memory entry per contact with exact channel identities (personal data). Private assistant content is never exposed in the list. Required scope: `leads:read` or `calls:read`.

Segment, 4 operations

create_segmentCreate segmentPOSTAgents: ask firstSave a lead filter as a reusable segment. Its matches are resolved live (dynamic, not a snapshot), see `list_segment_leads` and `assign_leads_to_campaign`.
delete_segmentDelete segmentDELETEAgents: ask firstDelete a segment. The leads it currently matches are not affected. Required scope: `segments:write` (the legacy `calls:write` scope is also accepted; keys without scope restrictions have full access).
get_segmentGet segmentGETAgents: ask firstRead a single segment. Required scope: `segments:read` (the legacy `calls:read` scope is also accepted; keys without scope restrictions have full access).
list_segmentsList segmentsGETAgents: ask firstList all segments (saved, dynamic lead filters) of the workspace. Required scope: `segments:read` (the legacy `calls:read` scope is also accepted; keys without scope restrictions have full access).

Call, 3 operations

create_callCreate callPOSTAgents: ask firstPlaces a real outbound phone call with an assistant (billed in workspace credits). The assistant must be active and have an outbound-capable number. Returns the call at once (status queued); read the result with get_call. Required scope: calls:write.
get_callGet callGETAgents: ask firstCall details including the transcript, the AI summary, provider-neutral failure guidance, the chronological event log, and a signed `recording_url` (valid 1 hour) if a recording exists.
list_callsList callsGETAgents: ask firstCalls in your workspace, newest first by creation time unless `sort=updated_at` is set. Failed outbound calls include a provider-neutral `failure` object with stable retry/action guidance.

Document, 3 operations

add_knowledge_base_documentAdd knowledge base documentPOSTAgents: ask firstAdd a document using exactly one of `content` (raw text; `name` required) or `url` (file URL, max 20 MB). Alternatively, provide `document_id` alone to retry an existing document synchronously, without creating or renaming one.
delete_knowledge_base_documentDelete knowledge base documentDELETEAgents: ask firstDeletes one knowledge-base document together with its indexed chunks and its stored original file. Assistants stop using its content right away.
list_knowledge_base_documentsList knowledge base documentsGETAgents: ask firstLists the documents of one knowledge base, newest first: uploaded files, raw text, file URLs, single web pages, crawled website pages and cloud drive files. Each document shows its source, indexing status and chunk count.

Mission, 3 operations

get_missionGet missionGETAgents: ask firstReturns one Milian Mission by id: prompt, schedule, status, published version and required toolkits. Required scope: routines:read.
list_missionsList missionsGETAgents: ask firstLists the workspace's Milian Missions (scheduled or triggered AI copilot tasks): name, prompt, schedule, status and the connected-app toolkits each one needs. Required scope: routines:read.
pause_missionPause missionPOSTAgents: ask firstPauses a Milian Mission and its trigger (versions and runs are kept). Only workspace owners and admins may pause. Required scope: routines:write.

Phone number, 3 operations

get_phone_numberGet phone numberGETAgents: ask firstReturns one workspace phone number: E.164 number, assigned assistant, SIP trunk, inbound and outbound directions and SMS capability. Required scope: phone_numbers:read.
list_phone_numbersList phone numbersGETAgents: ask firstPhone numbers of your workspace, newest first. Required scope: `phone_numbers:read` (keys without scope restrictions have full access).
search_available_phone_numbersSearch available phone numbersGETAgents: ask firstSearches purchasable phone numbers by country and type, with the workspace's retail price. Buying a number is not exposed here. The country and number type must be enabled by the platform admin. Required scope: phone_numbers:read.

Suppression list, 3 operations

add_suppression_entryAdd suppression entryPOSTAgents: ask firstRecords a marketing opt-out for a contact identified by contact ID, phone, or email. The current workspace consent mode determines whether the resulting scope is universal or limited to the originating channel. Required scope: `suppression:write`.
list_suppression_entriesList suppression entriesGETAgents: ask firstLists active marketing suppressions across voice, SMS, email, and supported messaging channels. Results expose the public contact identity, scope, origin channel, source, and consent mode captured at opt-out.
remove_suppression_entryRemove suppression entryDELETEAgents: ask firstRevokes one active suppression and appends an opt-in audit event; compliance history is not erased. `{id}` accepts the suppression entry ID, an E.164 phone number, or an email address.

Campaign lead, 2 operations

add_campaign_leadAdd campaign leadPOSTAgents: ask firstAdds one lead (phone required, normalized to E.164) to a campaign and returns it. A running campaign dials its leads in its scheduled waves (real, billed calls). Required scope: leads:write.
assign_leads_to_campaignAssign leads to campaignPOSTAgents: ask firstMoves unassigned contacts into a campaign, from a saved segment (segment_id) or explicit lead_ids (exactly one of the two). A running campaign dials them in its scheduled waves (real, billed calls). Required scope: leads:write.

Channel, 2 operations

list_audience_contact_channelsList audience contact channelsGETAgents: ask firstLists the manually manageable channel profiles linked to one Audience contact (personal data). Microsoft Teams profiles expose the visible connection name without internal account or routing identifiers. Required scope: `leads:read`.
replace_audience_contact_channelsReplace audience contact channelsPUTAgents: ask firstReplace the contact's manually managed channel profiles. For Microsoft Teams, provide the visible `connection_name`; it may be omitted only when exactly one Teams identity connection is active.

Dashboard, 2 operations

get_dashboardGet dashboardGETAgents: ask firstReturns one custom dashboard: name, sections and widgets layout. Required scope: dashboards:read. The workspace plan must include Custom Dashboards.
list_dashboardsList dashboardsGETAgents: ask firstReturns the workspace's custom dashboards in display order. Required scope: `dashboards:read` or the legacy `calls:read`. The workspace plan must include Custom dashboards.

Faq, 2 operations

create_knowledge_base_faqCreate knowledge base faqPOSTAgents: ask firstCreates human-approved FAQ knowledge and indexes it immediately. Required scope: `knowledge:write`.
list_knowledge_base_faqsList knowledge base faqsGETAgents: ask firstLists approved FAQ entries or the human-review knowledge-gap inbox. Required scope: `knowledge:read`. Beta Features must be enabled for the workspace.

Knowledge base faq, 2 operations

delete_knowledge_base_faqDelete knowledge base faqDELETEAgents: ask firstPermanently deletes one FAQ entry from a knowledge base; assistants stop using it. Required scope: knowledge:write.
get_knowledge_base_faqGet knowledge base faqGETAgents: ask firstReturns one FAQ entry of a knowledge base: question, answer, status (approved or a knowledge gap awaiting review) and source. Required scope: knowledge:read.

Mission run, 2 operations

list_mission_runsList mission runsGETAgents: ask firstLists a Milian Mission's recent runs, newest first: trigger, status, error, credits charged and timing. Run transcripts are kept for 10 days in the Famulor app. Required scope: routines:read.
run_missionRun missionPOSTAgents: ask firstStarts a background run of a Milian Mission now, outside its schedule. Spends workspace credits like a copilot chat turn. Returns the run (status running) at once; check list_mission_runs for the outcome. Required scope: routines:write.

Scheduled callback, 2 operations

cancel_scheduled_callbackCancel scheduled callbackPATCHAgents: ask firstCancels a pending or dialing scheduled callback so it will not be dialed. Required scope: calls:write.
list_scheduled_callbacksList scheduled callbacksGETAgents: ask firstCallbacks booked by the Schedule callback tool across voice, chat/messaging, and email. Due `pending` rows are dialed automatically by the platform dialer. Required scope: `calls:read`.

Webhook, 2 operations

get_webhook_subscriptionGet webhook subscriptionGETAgents: ask firstWorkspace owner/admin access is required. Required scope: `settings:read`. Signing secrets are never returned. Existing destinations remain unchanged.
list_webhook_subscriptionsList webhook subscriptionsGETAgents: ask firstWorkspace owner/admin access is required. Required scope: `settings:read`. Signing secrets are never returned. Existing destinations remain unchanged.

Webhook subscription, 2 operations

create_webhook_subscriptionCreate webhook subscriptionPOSTAgents: ask firstSubscribes your public HTTPS URL to call.completed, conversation.ended and booking events, signed with HMAC-SHA256 in X-Famulor-Signature using the secret you supply (never returned). Owner or admin; scope settings:write.
delete_webhook_subscriptionDelete webhook subscriptionDELETEAgents: ask firstDeletes a signed webhook subscription; deliveries to its URL stop. Answers deleted false when it was already gone. Workspace owner or admin only. Required scope: settings:write.

Widget, 2 operations

get_widget_connectorGet widget connectorGETAgents: ask firstReturns one website widget by id: public embed key, assistant, allowed origins, theme and verification settings. Required scope: assistants:read.
list_widget_connectorsList widget connectorsGETAgents: ask firstLists the workspace's website chat and voice widgets: public embed key, assistant, allowed origins, voice and chat switches, theme and visitor verification settings. Required scope: assistants:read.

Account, 1 operation

get_meGet meGETAgents: ask firstDescribes the calling API key: key metadata (name, scopes, expiry), the workspace, its active plan, numeric plan limits (-1 unlimited, 0 disabled) and plan feature toggles. Any valid key may call it.

Address, 1 operation

list_email_addressesList email addressesGETAgents: ask firstList addresses on the email channel, including the resolved domain and assigned assistant. Required scope: `assistants:read`. Requires workspace owner/admin.

Balance, 1 operation

get_balanceGet balanceGETAgents: ask firstMinutes and credits balance of the API key's workspace plus a summary of its active plan. Required scope: `billing:read` (keys without scope restrictions have full access).

Booking email template, 1 operation

get_booking_email_templatesGet booking email templatesGETAgents: ask firstReturns the standard text of the seven guest and host booking emails in the recipient language, plus the template variables custom subjects and bodies may use. Required scope: bookings:read.

Call recording, 1 operation

get_call_recordingGet call recordingGETAgents: ask firstReturns a time-limited signed download URL for a call's audio recording (with its expiry). Not found when the call has no recording. Required scope: calls:read.

Catalog, 1 operation

list_variable_sourcesList variable sourcesGETAgents: ask firstLists system variables, workspace lead attributes, and assistant variables available for campaign and WhatsApp template mappings. Required scope: `leads:read`.

Conversation reply, 1 operation

reply_to_conversationReply to conversationPOSTAgents: ask firstSends a human reply into an email or messaging conversation (delivered to the customer) and pauses AI replies there. Optional base64 attachments (up to 3 MB; messaging takes one file). Required scope: calls:write.

Dashboard analytics, 1 operation

get_dashboard_analyticsGet dashboard analyticsGETAgents: ask firstTenant-scoped call KPIs and deltas, zero-filled daily activity, breakdowns, assistant ranking, campaign progress, recent calls, and plan-gated module summaries. Date buckets follow the workspace timezone.

Delivery, 1 operation

list_campaign_deliveriesList campaign deliveriesGETAgents: ask firstLists primary and follow-up delivery attempts. Provider, room, carrier and lease identifiers are never exposed. Required scope: `campaigns:read`.

Email, 1 operation

get_email_conversationGet email conversationGETAgents: ask firstReturns every inbound and outbound turn in the email thread in chronological order, including stable thread and immediate-parent IDs, full plain-text bodies, reply states/errors, and attachment metadata.

Email sender, 1 operation

list_email_sendersList email sendersGETAgents: ask firstReturns the secrets-free sender choices that can be stored on a reusable `send_email` tool: automatic routing, workspace SMTP, platform mail, and verified sender addresses. Also returns workspace display-name and signature defaults.

History, 1 operation

list_historyList historyGETAgents: ask firstCalls and assistant email conversations in your workspace, globally sorted by latest activity. Failed outbound call rows include the provider-neutral `failure` object.

Language, 1 operation

list_languagesList languagesGETAgents: ask firstThe platform's assistant language catalog (ISO 639-1 or ISO 639-3 codes with English labels), managed by the platform admin. Use the codes for `primary_language` and `secondary_languages` on an assistant. Only enabled languages are returned.

Lookup, 1 operation

lookup_bookingsLookup bookingsPOSTAgents: ask firstFinds confirmed future bookings of the workspace by invitee identity, the same identity check assistants use before rescheduling or cancelling. Provide at least one of `phone`, `email`, `name`.

Messaging, 1 operation

get_messaging_conversationGet messaging conversationGETAgents: ask firstReturns up to 500 messages in chronological order with full text, attachment metadata, available voice-note transcripts, short image descriptions, and media links valid for one hour. Applies to chat channels and connected mailbox or support channels.

Messaging connector, 1 operation

list_messaging_connectorsList messaging connectorsGETAgents: ask firstList the workspace's messaging channel connections: chat apps, help desks and mailboxes. Each channel must be included in the workspace's plan; `meta.entitlements` shows which channels are available. Required scope: `assistants:read`.

Milian question, 1 operation

ask_milianAsk milianPOSTAgents: ask firstAsks Milian, Famulor's AI copilot, a read-only question about the workspace (answers, analysis, recommendations). Spends workspace credits on every call: set confirmed only after the user approved the cost. Do not retry automatically.

Model, 1 operation

list_modelsList modelsGETAgents: ask firstRequires the active workspace AI Safety & Control entitlement in addition to assistants:read. Returns compatible model choices filtered by workspace inference region. Whitelabel access alone does not grant catalog access.

Prompt template, 1 operation

list_prompt_templatesList prompt templatesGETAgents: ask firstSystem-prompt and conversational-flow templates visible to this workspace: platform templates (respecting global vs root visibility) plus workspace/reseller templates.

Search, 1 operation

search_knowledge_baseSearch knowledge basePOSTAgents: ask firstSearch document excerpts and exact table values with original cells and sources. Send either `query` (natural-language question) or `table_query` (exact lookup or aggregate on one sheet).

Slot, 1 operation

list_booking_slotsList booking slotsGETAgents: ask firstList the free start times of one active event type for a date range (at most 28 days; dates default to the workspace time zone). Past times and times after `bookable_until` are never returned.

Sms, 1 operation

send_smsSend smsPOSTAgents: ask firstSends a real SMS from a workspace number with outbound SMS enabled, billed in workspace credits. US 10DLC senders must be registered first. Required scope: phone_numbers:write.

Stats, 1 operation

get_campaign_statsGet campaign statsGETAgents: ask firstProgress and performance aggregates for a campaign: total and completed leads, calls made, total call duration, and delivery attempts by outcome.

Team member, 1 operation

list_team_membersList team membersGETAgents: ask firstList the current workspace team. Requires owner or admin access and settings:read.

Template, 1 operation

list_whatsapp_templatesList whatsapp templatesGETAgents: ask firstList HSM templates for the workspace. Required scope: `assistants:read`.

Transaction, 1 operation

list_transactionsList transactionsGETAgents: ask firstWorkspace balance ledger, newest first. `note` contains a customer-facing description; payment-provider IDs and internal reconciliation references are never returned.

Usage rate, 1 operation

get_usage_ratesGet usage ratesGETAgents: ask firstCurrent customer credit rates for the credential workspace, including fixed charges, token-based AI usage and variable charges. Requires billing:read or calls:read. A null credit amount means usage-based or variable pricing, never free usage.

Variable, 1 operation

get_assistant_variablesGet assistant variablesGETAgents: ask firstReturns the custom variable definitions of an assistant plus the inbound variable webhook URL. The webhook secret is never returned. Required scope: `assistants:read`.

View, 1 operation

list_history_viewsList history viewsGETAgents: ask firstList the workspace's saved History views, named filter sets shared by everyone in the workspace and shown as tabs on the History page. To list a view's conversations, pass its `id` as `view_id` to `list_history`. Required scope: `calls:read`.

Voice, 1 operation

list_voicesList voicesGETAgents: ask firstComplete paginated voice library for the active workspace. Voice IDs are provider-neutral local selectors; upstream identifiers and supplier names are never returned. Workspace-owned cloned voices use the same selector contract.

Workspace, 1 operation

list_workspacesList workspacesGETAgents: ask firstList the workspaces visible to the calling credential. For a user-owned API key or OAuth token: every workspace the key owner is a member of within this key's own brand (a reseller customer's key never sees a different reseller's workspaces).

Showing the first 120 of 130 endpoints. The rest are in the product, searchable from the workflow builder and from chat.

Use Famulor from Claude Code, Cursor or Codex

Connect Famulor in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Famulor 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 Famulor

Create your workspace
Sign up free. Nothing to install.
Connect Famulor once
Add your Famulor 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.