Skip to content
Scalekit Docs
Talk to an Engineer Dashboard

Connect AI agents to the Help Scout MCP server

Vendor MCP
Open markdown

The Help Scout MCP connector routes your AI agent's tool calls to Help Scout's own MCP server through Scalekit. Each user signs in to Help Scout once, and Scalekit stores and refreshes their tokens, so your agent never handles credentials. It comes with 54 tools.

Tools
54
What they doRead · write · destructive
54 · 0 · 054 read0 write0 destructive
Users sign in with
OAuth app
Your own Help Scout 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 Help Scout MCP connection

    In AgentKit > Connections, create a Help Scout 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

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

    Then enter the app's Client ID and Client Secret on the Help Scout 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 = 'helpscoutmcp'
    const identifier = 'user_123'
    // Generate an authorization link for the user
    const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
    console.log('Authorize Help Scout 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: 'helpscoutmcp_get_me',
    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
  • helpscoutmcp_docs_get_articleGet a single Docs article by ID (Requires linked Docs API key)Read-only

    Get Docs Article

    Get a single Docs article by ID (Requires linked Docs API key)

    Inputs

    articleIdstringrequired
    Article ID
    draftboolean
    Get draft version if true
  • helpscoutmcp_docs_get_siteGet a Docs site by ID (Requires linked Docs API key)Read-only

    Get Docs Site

    Get a Docs site by ID (Requires linked Docs API key)

    Inputs

    siteIdstringrequired
    Site ID
  • helpscoutmcp_docs_list_articlesList articles in a collection or in one category.Read-only

    List Docs Articles

    List articles in a collection or in one category. Pass exactly one of collectionId or categoryId; use docs_list_categories to find a category ID. status: "notpublished" returns articles that aren't live, including ones that were published before. A live article with unpublished edits has status "published" and hasDraft: true. Results are paginated. Pass page=1 (default) to start; check the response for total pages and request subsequent pages as needed. (Requires linked Docs API key)

    Inputs

    categoryIdstring
    Category ID
    collectionIdstring
    Collection ID
    orderstring
    No description.one of ascdesc
    pagenumber
    No description.
    pageSizeinteger
    Articles per page, up to 100. Defaults to 50.
    sortstring
    No description.one of numberstatusnamepopularitycreatedAtupdatedAt
    statusstring
    No description.one of allpublishednotpublished
  • helpscoutmcp_docs_list_categoriesList the categories in a Docs collection.Read-only

    List Docs Categories

    List the categories in a Docs collection. Pass a category ID to docs_list_articles to list only that category's articles. Results are paginated. Pass page=1 (default) to start; check the response for total pages and request subsequent pages as needed. (Requires linked Docs API key)

    Inputs

    collectionIdstringrequired
    Collection ID
    orderstring
    No description.one of ascdesc
    pagenumber
    No description.
    sortstring
    No description.one of numberordernamearticleCountcreatedAtupdatedAt
  • helpscoutmcp_docs_list_collectionsList collections in a Docs site.Read-only

    List Docs Collections

    List collections in a Docs site. Results are paginated. Pass page=1 (default) to start; check the response for total pages and request subsequent pages as needed. (Requires linked Docs API key)

    Inputs

    pagenumber
    No description.
    siteIdstring
    Filter by site ID
    visibilitystring
    public (customers can see it) or private (admin only). Defaults to all.one of allpublicprivate
  • helpscoutmcp_docs_list_sitesList Docs sites.Read-only

    List Docs Sites

    List Docs sites. Results are paginated. Pass page=1 (default) to start; check the response for total pages and request subsequent pages as needed. (Requires linked Docs API key)

    Inputs

    pagenumber
    Page number (default 1)
  • helpscoutmcp_docs_search_articlesSearch Docs articles by query.Read-only

    Search Docs Articles

    Search Docs articles by query. Results are paginated. Pass page=1 (default) to start; check the response for total pages and request subsequent pages as needed. An article is visible to customers only if it is published AND its collection is public: status: "published" alone still returns articles in private (admin-only) collections. To find content you can send a customer, pass status: "published" and visibility: "public". (Requires linked Docs API key)

    Inputs

    querystringrequired
    Search query string. Sent to the Help Scout Docs API article search endpoint: https://developer.helpscout.com/docs-api/articles/search/
    collectionIdstring
    Limit to a collection
    pagenumber
    No description.
    siteIdstring
    Limit to a site
    statusstring
    No description.one of allpublishednotpublished
    visibilitystring
    Visibility of the article's collection: public (customers can see it) or private (admin only). Defaults to all.one of allpublicprivate
  • helpscoutmcp_get_chat_reportGet the Chat report for a date range: live chat volume, wait and response times, and chat length.Read-only

    Get Chat Report

    Get the Chat report for a date range: live chat volume, wait and response times, and chat length. In `current`: volume.chatConversations is every chat conversation created in the range, including third-party chat integrations. Every other metric covers Beacon chats only. completedChats and missedChats count Beacon chat sessions, and one conversation can hold more than one, so completedChats + missedChats can be higher or lower than chatConversations. completedChats were accepted and then ended by either side. missedChats were never connected to a user, usually because no one was available, so the question moved to email. chatsPerDay is chatConversations divided by the days in the range. All times are in seconds, and are 0 (not null) when there were no chats. responses.waitTime is the average wait from the visitor starting the chat (or entering their email, when Beacon requires one) to a user accepting or being assigned it. Missed chats are excluded. responses.responseTime is the average time a customer waited for a user reply. Only the first user reply after each customer message counts, and chats with no user reply are excluded. resolutions.duration is the average time from a chat being accepted to it ending. resolutions.messagesPerChat is the average number of messages per chat. The top-level waitTime, responseTime, messagesPerChat and duration objects bucket chats by that metric, as a count and percent per range id. waitTime: 1=under 1 min, 2=1–2 min, 3=2–10 min, 4=10–20 min, 5=20 min or more. responseTime: 1=under 1 min, 2=1–2 min, 3=2–5 min, 4=5–10 min, 5=10 min or more. messagesPerChat: 1=under 5, 2=5–10, 3=10–20, 4=20–30, 5=30 or more. duration: 1=under 5 min, 2=5–10 min, 3=10–20 min, 4=20–30 min, 5=30 min or more. Ranges may not be listed in id order. Pass previousStart and previousEnd to also get `previous` (the same metrics for that range), `deltas` (percentage change per metric), and previousCount and previousPercent on each range. previousPercent is rounded down to a whole number, unlike percent. Without a comparison range, each bucket object's previousCount is 0, not missing. For one user's or team's chat metrics use get_user_chat_report.

    Inputs

    endstringrequired
    End date/time in ISO 8601 format
    startstringrequired
    Start date/time in ISO 8601 format
    foldersstring
    Comma-separated folder IDs
    mailboxesstring
    Comma-separated mailbox IDs
    officeHoursboolean
    Restrict metrics to office hours only. Ignored unless office hours are set up. Default false.
    previousEndstring
    End of the prior comparison period (ISO 8601)
    previousStartstring
    Start of the prior comparison period (ISO 8601)
    tagsstring
    Comma-separated tag IDs