Happy slice of mcp pizzaMCP.pizza

gongio-mcp

MCP.Pizza Chef: JustinBeckwith

Access your Gong.io calls, transcripts, users, and keyword trackers directly from AI apps like Claude Desktop. This server connects your Gong account to your AI assistant, letting you search calls by date, participant, or keywords, read full transcripts, and get AI-generated summaries with key points and action items. It requires a Gong admin API key and some technical setup using Node.js or Docker.

Communication
Other
Web/Research

Use This MCP server To

Find and read summaries of recent Gong calls Search Gong call transcripts for specific keywords Get detailed metadata about a specific call List all users and their profiles in my Gong workspace Filter calls by date range or participant Retrieve action items from call summaries Explore calls saved in Gong library folders

README

Gong MCP Server

npm version License: MIT

A cute red panda hitting a gong

An MCP (Model Context Protocol) server that provides access to your Gong.io data. Query calls, transcripts, users, keyword trackers, and more directly from Claude or any MCP-compatible client.

Tools Quick Reference

Tool Description
list_calls List calls with date/workspace filtering
get_call Get metadata for a specific call
get_call_summary AI summary: key points, topics, action items
get_call_transcript Full speaker-attributed transcript (paginated)
search_calls Rich call search — participant, customer, tracker, scope, duration, title, and more
search_calls_by_account Find calls involving a specific account/company by email domain
search_calls_by_opportunity Find calls linked to specific CRM Opportunities
search_transcripts Free-text keyword search across transcript sentences
get_trackers List keyword trackers (competitors, topics, etc.)
list_workspaces List workspaces and get IDs for use in other tools
list_library_folders List public call library folders
get_library_folder_calls Get calls saved in a specific library folder
get_user Get a specific user's profile
search_users Search/filter users by IDs or creation date
list_users List all workspace users

Response Size & Context Limits

search_calls can return a lot of data. Under the hood it:

  • Auto-paginates up to ~5000 calls (50 API pages) per query
  • Applies client-side filters (participant, customer, tracker, duration, etc.) after pagination
  • Returns a rich per-call format (metadata + summary + topics + participants) by default

Guardrails built in:

  • If the formatted output would exceed MAX_MCP_OUTPUT_LENGTH (default 50000 chars, configurable via env var), the tool automatically falls back to a compact table with a warning. You still get every call ID and title — drill in with get_call_summary on specific ones.
  • include: ["outline"] is expensive (~80KB per call). Avoid it in multi-call searches.
  • Tracker data is filtered to only show trackers matching your trackers filter (or non-zero trackers if no filter) — no more walls of (0x) noise.

If your query hits the output cap, narrow it:

  1. Tighten fromDateTime / toDateTime
  2. Add scope: "External" or scope: "Internal"
  3. Add minDuration: 600 to skip short no-shows
  4. Add customerName or trackers filter
  5. Drop heavy include options like outline

gongio-mcp FAQ

Can I use this to get summaries and transcripts of my Gong calls?
Yes — you can query call summaries, full transcripts, and metadata directly from your Gong.io account using this server.
Can I search Gong calls by participant or keywords?
Yes — it supports rich searches by participant, customer, keywords, date range, and more.
Which apps can use gongio-mcp?
It works with any MCP-compatible client, including Claude Desktop and Claude Code.
Do I need a Gong account or API key?
Yes — you must have Gong admin access to create API keys (Access Key and Secret) to use this server.
How hard is it to set up?
Setup requires developer skills like running Node.js or Docker and configuring environment variables.
Can I use this without coding?
No — it requires some technical setup such as running commands or Docker containers.
Does it support other AI models besides Claude?
Yes — MCP is model-agnostic, so you can use it with GPT-4, Gemini, Claude, etc.