Skip to content
Scalekit Docs
Talk to an Engineer Dashboard

Connect AI agents to the Vibe MCP server

Vendor MCP
Open markdown

The Vibe MCP connector routes your AI agent's tool calls to Vibe's own MCP server through Scalekit. Each user signs in to Vibe once, and Scalekit stores and refreshes their tokens, so your agent never handles credentials. It comes with 24 tools.

Tools
24
What they doRead · write · destructive
20 · 0 · 420 read0 write4 destructive
Users sign in with
OAuth app
Your own Vibe MCP server app

Setup

  1. Install the SDK

    Terminal window
    npm install @scalekit-sdk/node dotenv
  2. Set your credentials

    Add your Scalekit credentials to your .env file. Find values in app.scalekit.com > Developers > API Credentials.

    .env
    SCALEKIT_ENVIRONMENT_URL=<your-environment-url>
    SCALEKIT_CLIENT_ID=<your-client-id>
    SCALEKIT_CLIENT_SECRET=<your-client-secret>
  3. Create the Vibe MCP connection

    In AgentKit > Connections, create a Vibe MCP connection and copy its redirect URI. The name you give it is the connection_name your code passes. See Configure connections.

  4. Register an OAuth app

    Vibe MCP server connections use your own OAuth app. Register one with Vibe MCP server and add the redirect URI you copied.

    Then enter the app's Client ID and Client Secret on the Vibe MCP connection.

  5. Authorize a user and make your first call

    quickstart.mts
    import { ScalekitClient } from '@scalekit-sdk/node'
    import 'dotenv/config'
    import { createInterface } from 'node:readline/promises'
    const scalekit = new ScalekitClient(
    process.env.SCALEKIT_ENVIRONMENT_URL,
    process.env.SCALEKIT_CLIENT_ID,
    process.env.SCALEKIT_CLIENT_SECRET,
    )
    const actions = scalekit.actions
    const connector = 'vibemcp'
    const identifier = 'user_123'
    // Generate an authorization link for the user
    const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
    console.log('Authorize Vibe MCP:', link)
    const rl = createInterface({ input: process.stdin, output: process.stdout })
    await rl.question('Press Enter after authorizing...')
    rl.close()
    // Make your first call
    const result = await actions.executeTool({
    connector,
    identifier,
    toolName: 'vibemcp_list_advertisers',
    toolInput: {},
    })
    console.log(result)
    Terminal window
    npx tsx quickstart.mts

    Each user signs in once. See Authorize a user for the full flow and statuses.

Tools

Pass the exact name to execute_tool
Try in PlaygroundRequest a tool
  • vibemcp_get_activity_logReturns the change history (who changed what, and when) for the given records.Read-only

    Get Activity Log

    Returns the change history (who changed what, and when) for the given records. You only have access to change history from 2026-05-21 onward.

    Inputs

    modelsarrayrequired
    The entity types whose change history to fetch. By default the change history of each selected model's nested sub-records are included too; see `includeNestedRecords`.
    actionsstring
    Narrow to only creations and/or deletions of records (e.g. creatives added to or removed from a campaign). Omit to include all changes, including field updates.
    fromstring
    Filter changes on or after this date (inclusive), format "YYYY-MM-DD".
    includeNestedRecordsboolean
    Whether to also include changes to records nested under the selected models (e.g. CAMPAIGN also returns changes to its strategies, their creatives and targeting audiences, and its experiments). Keep the default (true): most edits to an entity are actually made on its nested records, so leaving them out will miss real changes. Only set false when the user explicitly wants changes to the named entity itself, excluding its children.default true
    limitstring
    Max number of changes to return, most recent first.default 25
    scopesstring
    Restrict results to records tied to these entities. Selected models unrelated to every listed entity are excluded. Omit to return history for ALL records of the selected models.
    sourcesstring
    Filter by one or more sources that made the change: APP (a user in the Vibe web app), AGENT (the in-app Vibe AI agent), PUBLIC_API (an external caller via the public API), MCP (an external AI agent or automation via the MCP), SYSTEM (an automated Vibe action with no human involved).
    tostring
    Filter changes on or before this date (inclusive), format "YYYY-MM-DD".
  • vibemcp_get_advertiser_detailsGet detailed advertiser information including website URL, industry, apps, and campaign count.Read-only

    Get Advertiser

    Get detailed advertiser information including website URL, industry, apps, and campaign count. Use this when the user asks about a specific advertiser beyond just its name. Returns an error if the advertiser ID does not exist or the caller lacks access.

    Inputs

    advertiserIdstringrequired
    The advertiser ID to look up
  • vibemcp_get_business_rulesLoad platform documentation for a knowledge domain.Read-only

    Get Business Rules

    Load platform documentation for a knowledge domain. Call when Vibe-specific rules or best practices apply to what you are about to do or explain.

    Inputs

    domainstringrequired
    The knowledge domain to retrieve: Use `getting_started` for Platform basics, account setup, and FAQ. Use `campaign_setup` for Campaign configuration, goals, targeting, creatives, and prerequisites. Use `pixel_tracking_reporting` for Pixel setup, conversion tracking, and reporting. Use `integrations` for Shopify, Google Analytics, Klaviyo, HubSpot, and other platform integrations. Use `billing_support` for Billing, payments, account management, and support topics. Use `campaign_optimization_playbook` for Optimization strategies, budget allocation, pacing, and performance best practices. Use `awareness_playbook` for Brand awareness, reach, frequency, and ad recall best practices.one of getting_startedcampaign_setuppixel_tracking_reportingintegrationsbilling_supportcampaign_optimization_playbookawareness_playbook
  • vibemcp_get_campaign_detailsGet campaign metadata including name, status, start/end dates, budget, and goal.Read-only

    Get Campaign

    Get campaign metadata including name, status, start/end dates, budget, and goal. Use this when the user asks about a specific campaign. Returns an error if the campaign ID does not exist or the caller lacks access. Includes the link to the campaign summary page on the Vibe dashboard, where the user can modify or publish a campaign.

    Inputs

    campaignIdstringrequired
    The campaign ID to look up
  • vibemcp_get_creative_detailsGet detailed creative information including approval status, duration, and whether it is used in an active campaign.Read-only

    Get Creative

    Get detailed creative information including approval status, duration, and whether it is used in an active campaign. Use this when the user asks about a specific creative. Returns an error if the creative ID does not exist or the caller lacks access.

    Inputs

    creativeIdstringrequired
    The video creative ID to look up
  • vibemcp_get_strategy_detailsGet detailed strategy information including targeting, bidding, and creatives.Read-only

    Get Strategy

    Get detailed strategy information including targeting, bidding, and creatives. Use this when the user asks about a specific strategy's configuration or targeting setup. Returns an error if the strategy ID does not exist or the caller lacks access.

    Inputs

    strategyIdstringrequired
    The strategy ID to look up
  • vibemcp_list_advertisersList advertisers for the current tenant.Read-only

    List Advertisers

    List advertisers for the current tenant. Use this to find advertiser names and IDs when the user asks about a specific advertiser or when you need to disambiguate.

    Inputs

    searchstring
    Search by advertiser name. Space-separated keywords return results that match any of them (better for broad searches). Keywords inside double quotes match the exact phrase (better for exact searches). Prefer a single call with multiple keywords over multiple separate tool calls.
  • vibemcp_list_campaignsList campaigns for the current tenant with optional filters, the most recently updated first.Read-only

    List Campaigns

    List campaigns for the current tenant with optional filters, the most recently updated first. Each campaign includes a displayStatus showing its current state as seen in the dashboard (e.g., Delivering, Paused, Upcoming, Completed, Draft, Archived, Blocked). Use the state filter for delivery states (DELIVERING, PAUSED, UPCOMING, COMPLETED, etc.). Use campaignStatus for lifecycle stages (DRAFT, PUBLISHED, ARCHIVED). Pass goal when the user mentions or implies a campaign objective (e.g., "lead campaign" → LEADS, "awareness campaign" → AWARENESS). Use publishedAfter/publishedBefore for publication date ranges.

    Inputs

    advertiserIdstring
    Filter by advertiser ID
    campaignStatusstring
    Filter by lifecycle status (DRAFT, PUBLISHED, ARCHIVED). Use DRAFT when user asks about draft campaigns. Most other user queries should use the state filter instead.
    goalstring
    Filter by campaign goal (e.g. AWARENESS, TRAFFIC, RETARGETING).
    limitstring
    Max results to return.default 50
    publishedAfterstring
    Filter campaigns published on or after this date in format "YYYY-MM-DD" (e.g. "2025-03-01").
    publishedBeforestring
    Filter campaigns published on or before this date in format "YYYY-MM-DD" (e.g. "2025-03-31").
    searchstring
    Search by campaign or advertiser name. Space-separated keywords return results that match any of them (better for broad searches). Keywords inside double quotes match the exact phrase (better for exact searches). Prefer a single call with multiple keywords over multiple separate tool calls.
    statestring
    Filter by delivery state (DELIVERING, PAUSED, UPCOMING, COMPLETED, BLOCKED, etc.). Use when the user asks about delivering, paused, upcoming, or completed campaigns.