Skip to content
Scalekit Docs
Talk to an Engineer Dashboard

Connect AI agents to the Render MCP server

Vendor MCP
Open markdown

The Render MCP connector routes your AI agent's tool calls to Render's own MCP server through Scalekit. Each user connects their own Render access token once, and Scalekit sends it with every call, so your agent never handles credentials. It comes with 48 tools.

Tools
48
What they doRead · write · destructive
32 · 10 · 632 read10 write6 destructive
Users sign in with

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 Render MCP connection

    In AgentKit > Connections, create a Render MCP connection. The name you give it is the connection_name your code passes. See Configure connections.

    Console steps with screenshots

    Register your Render API key with Scalekit so it can authenticate requests to the Render MCP server on behalf of your users. Render MCP uses API key authentication: the key is sent as a Bearer token.

    1. Create a Render API key

      • Log in to your Render dashboard.

      • Click your workspace name, then Account Settings.

      • Select API Keys from the settings sidebar.

        Render Account settings showing the API Keys panel with existing keys and a Create API Key button

      • Click Create API Key, give it a descriptive name (for example, Scalekit), and copy the key.

    2. Create a connection in Scalekit

      • In the Scalekit dashboard, go to AgentKit → Connections → Create Connection.
      • Search for Render MCP and click Create.
      • Note the Connection name — use this as connection_name in your code (for example, rendermcp).
    3. Add a connected account

      Connected accounts link a specific user identifier in your system to a Render API key. Add one in the dashboard to test. In production, each user adds their own through the authorization link: they enter their credentials on the page it opens.

      In the dashboard, to test

      • Open the connection and click the Connected Accounts tab → Add account.
      • Fill in Your User’s ID and API Key, then click Save.

      From your backend, if your app already has the credentials

      For example, when users enter them on a settings page in your app:

      import { ConnectorStatus } from '@scalekit-sdk/node'
      const authorizationDetails = {
      details: {
      case: 'staticAuth',
      value: { details: { token: process.env.USER_API_KEY } },
      },
      };
      let { connectedAccount } = await scalekit.actions.upsertConnectedAccount({
      connectionName: 'rendermcp',
      identifier: 'user_123',
      authorizationDetails,
      });
      // Make sure the account is ACTIVE before the first tool call.
      if (connectedAccount?.status !== ConnectorStatus.ACTIVE) {
      ({ connectedAccount } = await scalekit.actions.upsertConnectedAccount({
      connectionName: 'rendermcp',
      identifier: 'user_123',
      authorizationDetails,
      }));
      }
  4. 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 = 'rendermcp'
    const identifier = 'user_123'
    // Generate an authorization link for the user
    const { link } = await actions.getAuthorizationLink({ connectionName: connector, identifier })
    console.log('Authorize Render 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: 'rendermcp_get_selected_workspace',
    toolInput: {},
    })
    console.log(result)
    Terminal window
    npx tsx quickstart.mts

    Each user opens the link once and enters their Render MCP server credentials there. If your app already has a user's credentials, add the account from your backend instead, as the console steps above show. See Authorize a user for the full flow and statuses.

Tools

Pass the exact name to execute_tool
Try in PlaygroundRequest a tool
  • rendermcp_get_blueprintThis tool is in public alpha (early access).Read-only

    Get Blueprint details

    This tool is in public alpha (early access). Access to this tool and its contract are subject to change. Retrieve a Blueprint's settings and status, along with the IDs of the resources it manages. The tool returns the Blueprint detail object from the REST API. When autoSync is true, the Blueprint is configured to sync automatically on Git pushes to its configured branch. Automatic syncing requires approval of the Blueprint's first sync. When autoSync is false, syncing requires manual planning and approval. The Blueprint's status remains "created" until its first sync has been approved. After that approval, the status is "paused" whenever autoSync is false, regardless of the state of any manual sync. You can still use plan_blueprint and approve_blueprint while automatic syncing is paused. When autoSync is true and the first sync has been approved, the status reflects the latest sync. A status of "syncing" means that the sync is queued or running, and "error" means that it failed. The status can be "in_sync" when the latest sync has succeeded or when a plan is awaiting approval. Do not use the status alone to determine whether services are healthy or whether the infrastructure matches the current Blueprint file. To check for infrastructure changes, use plan_blueprint and follow its guidance for reviewing the plan with the user before approval. This tool does not create a plan or apply infrastructure changes.

    Inputs

    blueprintIdstringrequired
    This ID identifies the Blueprint to retrieve. Blueprint IDs start with exs-.
    workspaceIdstring
    The ID of the Render workspace to use. Reuse the workspaceId the user confirmed from list_workspaces.
  • rendermcp_get_deployRetrieve the details of a particular deploy for a particular service.Read-only

    Get deploy details

    Retrieve the details of a particular deploy for a particular service.

    Inputs

    deployIdstringrequired
    The ID of the deployment to retrieve
    serviceIdstringrequired
    The ID of the service to get deployments for
    workspaceIdstring
    The ID of the Render workspace to use. Reuse the workspaceId the user confirmed from list_workspaces.
  • rendermcp_get_flowGet a registered flow's name, workflow ID and version ID in the selected workspace.Read-only

    Get flow

    Get a registered flow's name, workflow ID and version ID in the selected workspace.

    Inputs

    flowIdstringrequired
    Flow ID
    workspaceIdstring
    The ID of the Render workspace to use. Reuse the workspaceId the user confirmed from list_workspaces.
  • rendermcp_get_flow_runInspect a flow run's status, error, error details and attempts (zero-indexed).Read-only

    Get flow run

    Inspect a flow run's status, error, error details and attempts (zero-indexed). A flow run gets a new attempt when it resumes after its process is interrupted; it replays completed steps rather than rerunning them. Errors are capped at 4 KiB each with errorTruncated flags. Inputs and results are omitted: use get_flow_run_results for results, list_flow_run_steps for steps, and list_logs with the workflow ID as resource and this ID as taskRun for logs. Poll this tool to wait for a run started with start_flow_run. Flow failure is data, not a tool failure.

    Inputs

    flowRunIdstringrequired
    Flow run ID
    workspaceIdstring
    The ID of the Render workspace to use. Reuse the workspaceId the user confirmed from list_workspaces.
  • rendermcp_get_flow_run_resultsRead a terminal flow run's results as paged JSON text, preserving arrays and integers exactly.Read-only

    Get flow run results

    Read a terminal flow run's results as paged JSON text, preserving arrays and integers exactly. Nonterminal runs return ready=false, not a final result. Concatenate content pages, passing nextOffset and sha256 on each follow-up; nextOffset=null marks the end. Do not infer success from ready=true: inspect status. Failed and canceled runs may have null or empty results.

    Inputs

    flowRunIdstringrequired
    Flow run ID
    limitnumber
    Maximum result bytes in this pagedefault 32768
    offsetnumber
    UTF-8 byte offset, use nextOffset from the preceding pagedefault 0
    sha256string
    Result hash from the first page; required for offsets greater than zero
    workspaceIdstring
    The ID of the Render workspace to use. Reuse the workspaceId the user confirmed from list_workspaces.
  • rendermcp_get_flow_run_stepGet one flow run step's status, result, and error.Read-only

    Get flow run step

    Get one flow run step's status, result, and error. The step's input is omitted. Errors are capped at 4 KiB each with errorTruncated flags. result is present only when the step succeeded; null means the step returned no value.

    Inputs

    flowRunIdstringrequired
    Flow run ID
    seqnumberrequired
    The step's seq from list_flow_run_steps
    stepKeystring
    The stepKey from list_flow_run_steps, used to distinguish steps with the same seq. Omit for a step recorded without a key
    workspaceIdstring
    The ID of the Render workspace to use. Reuse the workspaceId the user confirmed from list_workspaces.
  • rendermcp_get_key_valueRetrieve a Key Value instance by IDRead-only

    Get Key Value instance details

    Retrieve a Key Value instance by ID

    Inputs

    keyValueIdstringrequired
    The ID of the Key Value instance to retrieve
    workspaceIdstring
    The ID of the Render workspace to use. Reuse the workspaceId the user confirmed from list_workspaces.
  • rendermcp_get_metricsGet performance metrics for any Render resource (services, Postgres databases, key-value stores).Read-only

    Get resource metrics

    Get performance metrics for any Render resource (services, Postgres databases, key-value stores). Supports CPU usage/limits/targets, memory usage/limits/targets, service instance counts, HTTP request counts and response time metrics, bandwidth usage metrics, database active connection counts for debugging, capacity planning, and performance optimization. Returns time-series data with timestamps and values for the specified time range. HTTP metrics support filtering by host and path for more granular analysis. Limits and targets help understand resource constraints and autoscaling thresholds. Metrics may be empty if the metric is not valid for the given resource.

    Inputs

    metricTypesarrayrequired
    Which metrics to fetch. CPU usage/limits/targets, memory usage/limits/targets, and instance count metrics are available for all resources. HTTP request counts and response time metrics, and bandwidth usage metrics are only available for services. Active connection metrics are only available for databases and key-value stores. Limits show resource constraints, targets show autoscaling thresholds.
    resourceIdstringrequired
    The ID of the resource to get metrics for (service ID, Postgres ID, or key-value store ID)
    aggregateHttpRequestCountsBystring
    Field to aggregate HTTP request metrics by. Only supported for http_request_count metric. Options: host (aggregate by request host), statusCode (aggregate by HTTP status code). When not specified, returns total request counts.one of hoststatusCode
    cpuUsageAggregationMethodstring
    Method for aggregating metric values over time intervals. Only supported for CPU usage metrics. Options: AVG, MAX, MIN. Defaults to AVG.one of AVGMAXMINdefault AVG
    endTimestring
    End time for metrics query in RFC3339 format (e.g., '2024-01-01T13:00:00Z'). Defaults to the current time. The end time must be within the last 30 days.
    httpHoststring
    Filter HTTP metrics to specific request hosts. Supported for http_request_count and http_latency metrics. Example: 'api.example.com' or 'myapp.render.com'. When not specified, includes all hosts.
    httpLatencyQuantilenumber
    The quantile/percentile of HTTP latency to fetch. Only supported for http_latency metric. Common values: 0.5 (median), 0.95 (95th percentile), 0.99 (99th percentile). Defaults to 0.95 if not specified.default 0.95
    httpPathstring
    Filter HTTP metrics to specific request paths. Supported for http_request_count and http_latency metrics. Example: '/api/users' or '/health'. When not specified, includes all paths.
    resolutionnumber
    Time resolution for data points in seconds. Lower values provide more granular data. Higher values provide more aggregated data points. API defaults to 60 seconds if not provided. There is a limit to the number of data points that can be returned, after which the metrics endpoint will return a 500. If you are getting a 500, try reducing granularity (increasing the value of resolution).
    startTimestring
    Start time for metrics query in RFC3339 format (e.g., '2024-01-01T12:00:00Z'). Defaults to 1 hour ago. The start time must be within the last 30 days.
    workspaceIdstring
    The ID of the Render workspace to use. Reuse the workspaceId the user confirmed from list_workspaces.