MCP App Store
by KeenEthics
Geckoboard icon

Geckoboard MCP Server for Claude and ChatGPT

By Geckoboard

Overview

Access accurate, up-to-date business metrics from Geckoboard directly in Claude. Query data from Zendesk, HubSpot, Pipedrive, Intercom, Aircall, Google Analytics, and more without exporting records or calculating metrics by hand. With the Geckoboard connector, Claude can investigate changes, draft reports, trigger workflows, and build data-powered tools.

How to connect

  1. Claude: Open the Geckoboard listing in the Claude connectors directory and choose Connect. Claude listing Geckoboard exposes 9 tools here.
  2. ChatGPT: Open the Geckoboard listing in the ChatGPT app directory and choose Connect. ChatGPT listing Geckoboard exposes 9 tools here.

Tools

get_data_list

ChatGPT
Fetch the full detail for a single data list: available fields, default fields, required filters, filters, and sorting fields in addition to the summary fields. Use this to gather the field and filter IDs needed to build a query_data_lists call. Filter fields carry their constraints: overrideFilterOperators (the only operators valid for that field, when set), canBeNull, filterValueLimit, filterDependencies, and incompatibleWith. Every field listed in requiredFilters MUST be supplied as a filter for the query to execute. defaultSortBy, when set, is the natural sort for the data list and is applied when query_data_lists omits sortBy. sourceItemURL, when set, is a {{field.id}} template for linking each row back to its upstream item.

get_metric

ChatGPT
Fetch the full detail for a single metric: required filters, filters, and dimensions in addition to the summary fields. Use this to gather the filter and dimension IDs needed to build a query_metrics call. Filter fields carry their constraints: overrideFilterOperators (the only operators valid for that field, when set), canBeNull, filterValueLimit, filterDependencies, and incompatibleWith. Every field listed in requiredFilters MUST be supplied as a filter for the query to execute. aggregateSupport, when set, lists in aggregateOperators the operators valid for the metric aggregate in query_metrics, each with this metric’s own wording for it where it has one; when null the metric does not support aggregation. supportedBuckets lists the time units valid for a TIME dimension, supportedCustomTimeUnits and supportedAllTimeBuckets cover the integration-wide units for custom timespans and all-time bucketing, minTimespan / maxTimespan bound the timespan range for time-series queries, supportsFutureTimespans says whether direction FUTURE is allowed, recommendedDimension names a sensible default dimension, and relatedDataList links to the data list behind the metric for fetching raw rows.

list_connections

ChatGPT
List the integration connections available to query for data. Returns the connection ID, name, and integration type for each connected integration. You do not need this to browse a catalogue: list_metrics and list_data_lists accept an integration name and resolve the connection themselves. Reach for this when the user asks what is connected, when you need to see connection names to tell several connections of the same integration apart, or when an integration name did not match anything.

list_data_lists

ChatGPT
List the data lists (tables/feeds) available for an integration connection. Identify the connection either by integration — an integration name as loosely as the user said it, which is resolved for you and saves a list_connections call — or by connectionId plus integrationSlug when an earlier call already gave you both. Pass searchTerm to narrow the catalogue to the data lists you're after — only omit it when the user genuinely wants to browse. Returns a slim payload per data list (ID, display name, description), the integration capabilities that gate certain query features, and the connection the catalogue came from. This is the catalogue step: to get actual rows, pick a data list here, call get_data_list for its full field, filter, and sorting details, then call query_data_lists.

list_field_values

ChatGPT
List the possible values of a COLLECTION or ENUM field so they can be used in IS_IN / IS_NOT_IN filters. Use a field ID from get_metric or get_data_list (including requiredFilters). For fields with many values, pass a searchTerm to narrow the results. Returns the value ID and display name for each option.

list_metrics

ChatGPT
List the metrics available for an integration connection. Identify the connection either by integration — an integration name as loosely as the user said it, which is resolved for you and saves a list_connections call — or by connectionId plus integrationSlug when an earlier call already gave you both. Pass searchTerm to narrow the catalogue to the metrics you're after — only omit it when the user genuinely wants to browse the whole catalogue. Returns a slim payload per metric (ID, display name, data type, description, support flags), the integration capabilities that gate certain query features, and the connection the catalogue came from. This is the catalogue step: to get actual values, pick a metric here, call get_metric for its full filter and dimension details, then call query_metrics.

list_scope_values

ChatGPT
List the possible values for a scope on a connection. Use the scope IDs returned by list_metrics or list_data_lists. Returns the value ID and display name for each option.

query_data_lists

ChatGPT
Run a data list query and return the actual underlying rows. This is the tool that answers questions like "show me the deals/tickets/orders that match X" — it executes against the live integration data and returns tabular data with typed columns and rows. Use data list IDs and field IDs gathered from list_data_lists and get_data_list; supply every field listed in the data list's requiredFilters.

query_metrics

ChatGPT
Run a metric query and return the actual aggregated values. This is the tool that answers questions like "how many X yesterday" or "X broken down by Y" — it executes against the live integration data and returns tabular data with typed columns and rows, optionally grouped by dimensions and bucketed over time. Use metric IDs, filter IDs, and dimension IDs gathered from list_metrics and get_metric; supply every field listed in the metric's requiredFilters.

get_data_list

Claude

get_metric

Claude

list_connections

Claude

list_data_lists

Claude

list_field_values

Claude

list_metrics

Claude

list_scope_values

Claude

query_data_lists

Claude

query_metrics

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/geckoboard"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/geckoboard)