Beginner-friendly Votel MCP guide

Your AI assistant can work inside Votel.

Think of MCP as the secure bridge between your AI assistant and your Votel account. You describe what you want in plain English; the assistant can use approved Votel tools to help build it.

318Live tools
23Categories
0Code needed to start
Live inventory verified September 17, 2026. Tool counts change as Votel ships updates.
How the connection works
Your preferred AI clientYou chat in the tool you already use.
ClaudeChatGPTCursorVS Code
Votel MCP serverSecurely describes the available tools and carries approved requests.
OAuthAPI key
Your Votel workspaceContacts, agents, workflows, websites, settings, analytics, and more.
Audit the accountBuild draftsUpdate recordsRun reportsPublish with approval
Start here · no technical background needed

What the heck is an MCP?

MCP gives your AI assistant hands inside Votel.

Normally, Claude or ChatGPT can tell you how to do something. Once connected to Votel’s MCP, it can use approved Votel tools to inspect your account, create drafts, update records, build assets, and report back—while you stay in the conversation.

Without MCP

The AI gives you instructions

It tells you which menus to open. You still click through Votel and build everything yourself.

With Votel MCP

The AI can do the work

You describe the result. The assistant chooses approved tools and Votel performs each authorized action.

What does MCP stand for?

Model Context Protocol. It is an open standard that lets an AI application discover tools and data offered by another system. Your AI app is the client, Votel supplies the MCP server, and each named Votel tool is an action the assistant can request.

The important part

Why is this so powerful?

You do not need to learn 318 tool names. You explain the business outcome in ordinary language. Your assistant works out which tools it needs, asks for missing information, and can complete a multi-step job across several parts of Votel.

1

Describe the result

“Build a lead-capture system for my roofing client.”

2

Review the plan

The assistant identifies the form, fields, pipeline, workflow, calendar, and dashboard it needs.

3

Approve the work

Votel creates the drafts and the assistant verifies what was actually saved.

One prompt can become an entire customer journey.That is the difference between calling one endpoint and giving an assistant a toolbox.
MCP tool vs. API endpoint

A developer calls a REST API endpoint directly in code. An MCP tool has a name, description, and structured inputs that an AI assistant can understand and select. Votel offers both: its REST API lives at https://app.votel.ai/api, while its remote MCP server lives at https://app.votel.ai/mcp.

A real-world agency example

What happens after you paste a prompt?

Imagine you are onboarding a roofing company. You want every website lead captured, organized, followed up with, and offered a booking link. Here is what one MCP-assisted conversation could look like.

Prompt pasted into Claude Code, ChatGPT, or another connected client
Set up a draft lead-capture journey for Summit Roofing. Create the fields needed for service type and property ZIP code, a “Request a roof inspection” form, a New Lead pipeline stage, an inactive follow-up workflow, a roof-inspection calendar, and a dashboard showing new leads and booked inspections. Show me the plan first. Do not publish, activate, send messages, or buy anything without my approval.
01
Assistant reads first

It checks the workspace

It looks for existing fields, pipelines, forms, calendars, and naming conventions so it does not create unnecessary duplicates.

list_custom_fields · list_pipelines · get_form_context · get_scheduling_context
02
You approve the plan

It creates connected drafts

After approval, it creates the missing CRM structure, form, booking resource, inactive workflow, and dashboard.

create_custom_field · create_form_from_plan · create_bookable · create_dashboard
03
Assistant checks its work

It verifies the saved result

It confirms IDs, field mappings, inactive status, launch blockers, and anything that still requires a human decision.

preview_form_visibility · get_dashboard · verify_artifact
VOTEL WORKSPACESummit Roofing · Draft customer journeyNot published
CRM2 fields + New Lead stageService type · Property ZIP
FORMRoof inspection requestDraft · fields connected
CALENDARRoof inspectionsDraft availability ready
WORKFLOWNew-lead follow-upInactive · awaiting copy approval
DASHBOARDLead & booking overview2 widgets configured
ASSISTANT REPORT5 workspace areas configured · 0 items publishedNext decision: approve messaging copy and test with one sample lead.

The exact tools may differ. The important idea is that you ask for the business outcome, review the plan, approve the work, and receive a verified result—not a list of endpoints to decipher.

About five minutes for most remote connections

Connect the AI tool you already use

For most beginners, remote OAuth is the cleanest option: paste one server URL, sign in to Votel, and approve access. Local API-key setup is useful for developer environments or clients that connect through a local command.

Recommended for beginners

Connect with remote OAuth

The wording varies by AI client, but the basic process is the same. You will not need to copy or expose an API key.

  1. 01
    Open your AI tool’s connection settings.

    Look for a menu named MCP, Connectors, Integrations, or Developer settings.

  2. 02
    Add a remote MCP server.

    Choose the option to add a custom or remote server. Name it Votel when your client asks for a name.

  3. 03
    Paste Votel’s MCP server address.
    https://app.votel.ai/mcp
  4. 04
    Sign in to Votel and approve access.

    Your browser opens Votel’s authorization screen. Confirm the correct Votel account and complete the connection.

  5. 05
    Test the connection without changing anything.

    Ask: “List my Votel websites. Do not create, edit, or publish anything.” If the assistant returns information from your workspace, you are connected.

Client menus changeView Votel’s current setup instructions for your AI tool
Client Remote OAuth Local API key Good fit
Claude.ai Yes, paid plans Beginners who want a familiar chat interface
ChatGPT Yes, supported paid accounts Everyday planning, building, and workspace management
Claude Code Yes Yes Terminal-based, multi-step implementation work
Claude Desktop Yes, paid plans Yes Desktop chat with either connection style
Cursor / Windsurf Yes Yes Website and app work tied to a codebase
VS Code / Cline VS Code supports both; Cline supports local Yes Developer workflows and project-scoped setup
Using a local API-key connection? This is mainly for developer environments and requires Node.js 18+ plus a Votel secret API key. Follow Votel’s official MCP Setup Guide for the current command and client-specific configuration. Never paste the secret key into a public chat, screenshot, or repository.
Pick the job, not the endpoint

What can it actually do?

The value is not “having 318 endpoints.” The value is giving your assistant enough approved capabilities to complete a useful outcome across several parts of the workspace.

Understand before changing

Audit a workspace

Inspect agents, workflows, forms, channels, integrations, domains, teams, and missing setup without editing anything.

Audit this workspace. Make no changes. Give me a prioritized plan and cite the exact configuration behind each finding.
Build a customer journey

Create connected assets

Draft a contact field, form, pipeline, AI agent, workflow, calendar, and dashboard around one business outcome.

Plan a lead-to-booking journey. Show me the assets and dependencies before creating drafts.
Run daily operations

Work the CRM and inbox

Find contacts, update fields, add notes, assign conversations, create tasks, move pipeline stages, and schedule follow-up.

Find new leads without a next task. Propose the task list first, then wait for approval.
Create and publish

Launch marketing assets

Build websites, funnels, forms, email designs, social posts, images, branded tracking links, and dashboards.

Create a draft campaign package using our brand kit. Do not publish or spend money.
See what happened

Analyze performance

Query calls, messages, form submissions, workflow runs, bookings, clicks, ad results, activities, and custom dashboards.

Summarize the last 30 days by lead source, booked appointment, call outcome, and revenue.
Scale agency delivery

Set up client workspaces

Create sub-accounts, configure white-label settings and SaaS plans, invite teammates, and apply repeatable operating patterns.

Review our packages and propose a new client workspace plan. Do not create the account until approved.
Automate repeatable work

Schedule assistants and workers

Create scheduled tasks, specialist subagents, contact workers, messaging channels, and connected external MCP servers.

Draft a weekly account-health task that saves its report to the Library. Keep it disabled.
Get current answers

Use docs and the web

Search Votel help articles, read changelog entries, research the live web, and extract pages before acting.

Check current Votel documentation for this feature, then explain the safest implementation path.
FROM MCP DATA TO A CLIENT-READY DELIVERABLETurn your account into a Votel Snapshot Blueprint.Use Fer’s documentation engine to organize your MCP data into grouped inventories, relationship maps, an AI System view, account findings, HTML, CSV, and Excel.Explore the Snapshot Blueprint →

Want to understand the underlying process first? Read the self-service account documentation guide →

All 23 tool categories

These counts come from Votel’s live MCP Tools screen on September 17, 2026. Search by outcome, category, or example tool name.

23 of 23 categories shown

CRM

64

Contacts, custom fields, pipelines, tasks, projects, conversations, notes, tags, users, layouts, opt-outs, and activity.

create_contact · list_contacts · move_lead_stage · create_task · update_contact_layout

Agents

19

Create, configure, audit, template, activate, and troubleshoot AI agents, knowledge, inline tools, widgets, and workflows.

create_ai_agent · audit_ai_agent_config · configure_agent_knowledge · create_agent_workflow

Social

10

Connect accounts, read platform rules and brand voice, then draft, schedule, publish, retry, or delete posts.

connect_social_account · create_social_post · get_social_posting_context

Images

6

Generate on-brand images, inspect models and aspect ratios, create links, and organize results in the Media Library.

generate_images · get_image_generation_context · organize_images

Branding

4

Create a brand kit from a website or supplied guidelines, then read, update, or retire it.

create_ai_brand_kit · get_branding_context · update_brand_kit

Forms

12

Plan complete forms, add and edit elements, preview conditions, connect resources, check readiness, and publish.

create_form_from_plan · preview_form_visibility · publish_form · update_form_element

Scheduling

20

Manage booking links, services, staff, availability, bookings, outcomes, rescheduling, cancellations, counts, and context.

create_bookable · create_booking · find_bookings · update_scheduling_staff

Calls

7

Place an AI phone call, check progress, search history, read outcomes and transcripts, obtain recordings, and report stats.

make_ai_phone_call · get_call_transcript · get_call_stats · search_calls

Messaging

5

Inspect senders and rules, send or schedule email and SMS, review the queue, and cancel pending messages.

get_messaging_context · send_email · send_sms · cancel_scheduled_message

Email

7

Build and edit email designs, inspect layouts and variables, find stock photography, and import selected images.

create_email_design · update_email_design · search_stock_images

Click Tracking & Ads

28

Create tracking links and domains, measure ROI, inspect ad accounts, build audiences and ads, validate specs, and control spend state.

create_tracking_campaign · query_clicktrack_stats · validate_ad_spec · set_ad_status

Affiliates

14

Manage affiliates, applications, terms, commissions, balances, tiers, payout batches, approvals, and paid status.

create_affiliate · review_commission · create_payout_batch · list_affiliate_payouts

Analytics

2

Discover available metrics and filters, then answer totals, averages, grouped, and time-series questions with real workspace data.

get_analytics_context · query_analytics

Activity

3

Inspect available activity logs, query their rows, and open the full details behind a call, message, submission, or run.

get_activity_log_context · query_activity_logs · get_activity_detail

Dashboards

4

Discover widget types and grid rules, build a dashboard, read its layout, and update widgets or placement.

get_dashboard_builder_context · create_dashboard · update_dashboard

Websites

23

Create and publish CDN websites, read and patch files, manage versions, domains, DNS, cache, storage, status, and traffic analytics.

create_website · publish_site · patch_website_file · verify_domain · restore_website_file

Funnels

5

Read the design blueprint, build complete funnels, inspect launch blockers, connect Stripe, and update pages, offers, and automations.

create_funnel · get_funnel_status · update_funnel · list_stripe_integrations

Web

3

Search current outside information, research across sources, and extract the full text of specific pages.

web_search · web_research · web_extract

Sub-Accounts

2

List available packages and provision a new client sub-account on the approved package.

list_packages · create_sub_account

Settings

36

Manage teams, roles, invitations, integrations, domains, numbers, global variables, SaaS branding, plans, routing, and tests.

get_team_context · connect_integration · update_saas_branding · buy_phone_numbers

Assistant

34

Configure Votel’s own assistant with messaging channels, external MCP connectors, scheduled tasks, subagents, workers, audits, and verification.

run_account_audit · create_scheduled_task · create_subagent · create_worker · verify_artifact

Library

6

Read and save workspace files, inspect snapshots, and browse scheduled-task outputs and run history.

list_library_files · save_library_file · list_snapshots · get_scheduled_task_runs

Help & Docs

4

Search Votel help documentation, read a source article, and inspect product changelog entries instead of relying on memory.

search_help_docs · read_help_article · list_changelog · get_changelog_entry
Why this page does not freeze all 318 tool definitions: names, inputs, and descriptions can change. The always-current list is inside your Votel account at Settings → API / MCP → MCP Tools. This guide explains the use cases; the live directory remains the technical source of truth.

Beginner-safe prompts to copy

Good MCP requests define the outcome, the scope, what the assistant must not do, and where approval is required.

Read-only inventory

Audit this Votel workspace. List what exists for CRM, agents, workflows, forms, websites, scheduling, email, phone, integrations, branding, dashboards, and sub-accounts. Do not create, edit, send, publish, purchase, invite, activate, or delete anything. Give me a prioritized plan.

Draft before building

I want to build [OUTCOME] for [BUSINESS]. First inspect the relevant workspace context and documentation. Show me the proposed assets, dependencies, fields, channels, and test plan. Wait for approval before creating drafts, and keep everything unpublished and inactive.

CRM cleanup

Review my contacts, fields, pipelines, tags, views, and open tasks. Identify duplicates, unused structure, missing follow-up, and risky changes. Do not delete, merge, opt out, or overwrite records. Return a cleanup plan with counts and examples.

Verify the work

Verify the artifact we just built against its saved configuration and intended customer journey. Check prerequisites, inactive or unpublished state, field mappings, routes, permissions, and likely failure points. Do not activate or publish it.

Use powerful tools with clear guardrails

A connected assistant can do real work. Treat it like a capable operator: give it a defined job, limited authority, and explicit approval points.

Start with context tools

Ask for get_*_context, list_*, audits, status checks, and previews before write actions.

Keep first builds inactive

Draft forms, agents, workflows, ads, websites, and scheduled tasks before publishing or activation.

Require approval for consequences

Sending messages, placing calls, inviting users, publishing, spending money, buying numbers, and creating payouts need an explicit checkpoint.

Be careful with destructive tools

Deletion, domain removal, number release, key changes, and permanent publishing actions can be difficult or impossible to undo.

Protect credentials

Prefer OAuth where supported. Never paste a secret API key into a public chat, screenshot, website, or repository.

Verify after every build

Use audits, status tools, previews, test records, and verify_artifact before a customer sees the result.

Use the current documentation

Last reviewed September 17, 2026.

The live Votel account showed 318 tools across 23 categories. Votel’s public setup guide still describes “290+” tools, which shows why a live inventory link matters more than a permanently copied list.

All Things Votel is independent and not operated by or affiliated with Votel.ai. Tool availability, client compatibility, and permissions can change. Confirm consequential actions in your own workspace before approving them.