Overview

The Featurebase MCP server lets AI tools and assistants like Claude, ChatGPT, and Cursor securely connect to your Featurebase workspace.

Once connected, AI agents can read, create, and update data across Featurebase - no custom integration work required.

What is Featurebase MCP?

MCP (Model Context Protocol) is an open standard for connecting AI tools to external systems, so any MCP-compatible client can talk to Featurebase out of the box.

Featurebase MCP uses our public API to give AI tools a structured set of actions you can run through conversation:

• Easy setup - Connect with OAuth in a few clicks and start managing your Workspace through chat

• Works with MCP-compatible clients - Claude, ChatGPT, Cursor, VS Code, Claude Code, Codex, and other clients that support remote MCP servers

• Broad Workspace coverage - Work with feedback, support Conversations, customer data, Updates, Help Center content, Reports, Surveys, and webhooks

• Secure by design - AI tools authenticate as a Featurebase teammate and inherit that teammate's existing permissions

One connection with separate permissions

Featurebase MCP gives your AI tool read-only and write/delete tools through one secure connection. Your AI client can present these tools separately, so you can allow a tool, require confirmation before it runs, or turn it off.

• Read-only tools - Search, list, and retrieve Featurebase data without changing it

• Write/delete tools - Create, update, publish, reply to, or delete Featurebase data

Your Featurebase Workspace permissions still apply. The connected AI tool only receives tools and data that the authenticated teammate can access.

What you can do with Featurebase MCP

Read and analyze

• Feedback & Roadmaps - Search and review feedback posts, comments, votes, and voters

• Help Desk - Review Conversations, Tickets, customer history, categories, statuses, and attributes

• Help Center - Search and read articles, collections, Help Centers, and redirect rules

• Users directory - Look up People, Companies, and customer data

• Reports and Surveys - Query reporting data and analyze Survey responses

• Updates and Workspace configuration - Review Updates, audit logs, boards, statuses, tags, brands, teams, and webhooks

Create and manage

• Feedback & Roadmaps - Create and update posts and comments, and manage voters

• Help Desk - Create and update Conversations and Tickets, send replies, manage participants and tags, and redact Conversation content

• Help Center - Create, edit, publish, and delete articles, collections, and redirect rules

• Updates - Create, edit, publish, unpublish, and delete Updates

• Users directory - Create, update, block, unblock, and delete contact and Company records

• Webhooks - Create, update, remove, and refresh signing secrets

Available tools depend on your Workspace permissions and the settings in your AI client. For the complete API surface behind the MCP, see our API reference.

Getting started

Featurebase MCP is available on all paid plans. Before connecting, ensure you have an active Featurebase Workspace, permission to manage API access, and an MCP-compatible AI client.

Connect Featurebase with your AI tool:

1. Go to Featurebase Settings → MCP

2. Choose your AI tool and follow its install instructions

3. Complete the Featurebase OAuth flow to connect your Workspace

4. Open the connection's tool permissions in your AI client and review the read-only and write/delete tools

Your AI client now has one Featurebase connection. You can disconnect it at any time under Settings → MCP → Connected clients.

Example use cases

One Featurebase MCP connection can support research, content, support, and administrative workflows.

Research and reporting

• Analyze feedback trends - Identify recurring themes, sentiment, and feature interest across incoming feedback

• Surface Conversations by customer, topic, or conversation attribute - Find open issues for a specific customer or summarize Conversations about a product area, problem, or time period

• Analyze Survey responses - Find patterns and sentiment across open-ended Survey responses

• Generate recurring digests - Create weekly or monthly summaries of feedback, Conversations, support topics, or customer segments

• Summarize customer history - Review a customer's previous Conversations, feedback, and feature requests before a meeting or reply

Customer support

• Draft support replies - Pull the complete Conversation context and draft a response grounded in the customer's history

• Prepare escalation briefs - Summarize the issue, previous replies, attempted solutions, and unresolved questions before handing a Conversation to another teammate

• Identify recurring support questions - Group similar Conversations to uncover common problems, confusing workflows, and documentation gaps

• Chat with your Help Center - Ask questions across your documentation and receive answers grounded in your published articles

• Draft and update Help Center articles - Create articles, refresh outdated content, or restyle existing documentation through chat

• Audit documentation for gaps - Compare completed Featurebase posts or product changes with existing Help Center content, then apply approved updates

• Generate articles from support trends - Find recurring questions in Conversations and turn the reviewed findings into relevant Help Center content

• Keep Help Center articles current after releases - Identify affected articles, review recommended changes, and update approved pages

• Turn Updates into Help Center articles - Expand a release announcement into durable setup, reference, or troubleshooting documentation

Improve Fibi's knowledge

Use the Training Data tools to manage Q&A snippets and files. Your agent can turn reviewed support conversations or product changes into proposed knowledge updates, then ask the read-only oracle whether to create a Q&A, update an existing entry, or review a related article or file.

Your AI tool controls the schedule and approval steps. For setup instructions and example prompts, see Build a self-improving Fibi workflow.

Feedback operations

• Turn call transcripts into feedback posts - Convert reviewed sales or support call notes into structured feedback posts

• Triage incoming feedback - Cluster and prioritize posts, then apply approved changes

• Maintain feedback at scale - Create and update posts and comments, and manage voters

• Turn research into structured feedback - Convert reviewed interview notes, research summaries, or imported feedback into organized posts

Updates communication

• Draft Updates from release notes - Turn release notes into clear, properly categorized Updates

• Create Updates from completed work - Combine completed Featurebase posts with Linear or Jira issues, release notes, or codebase context available to your AI tool

• Turn Help Center articles into Updates - Condense detailed documentation into a concise release announcement

Workspace administration

• Manage customer records in bulk - Create or update contact and Company records from a reviewed list

• Keep support work organized - Manage Conversation tags, Ticket details, and Conversation participants

• Manage webhooks - Create webhooks, refresh signing secrets, or remove obsolete connections

FAQ