ReleasePad Launches MCP Server to Help AI Agents Draft Product Changelogs

Changelog tool ReleasePad launched a Model Context Protocol server on September 14, 2026, enabling AI coding agents like Claude, Codex, and Grok to draft, publish, and analyze product release notes directly from developer conversations. The integration connects via OAuth without requiring API keys, targeting a task many software teams skip.

Software teams ship new features every week, but announcements often stall before reaching customers. Release notes represent the step most teams skip: the feature ships, the announcement waits, and customers never hear about it. ReleasePad aims to close that gap by integrating directly with the tools developers already use during coding sessions.

How the ReleasePad MCP Server Connects AI Agents to Product Updates

The newly launched Model Context Protocol server allows AI assistants to read what a team merged into their codebase, draft a release note matching the team’s voice, categorize the update, attach images, and schedule publication. Setup requires a single connection through OAuth, eliminating the need to manage API keys.

Developers can prompt their assistant in natural language, using instructions such as Look at what we merged this week, draft a release note for customers, and leave it as a draft for me. Beyond drafting new entries, assistants can pull up earlier posts to edit or reference, or analyze past performance metrics. Despite this automation, safety guardrails remain intact: Every draft wait for a person to approve it; nothing is published until someone presses the button, according to the source.

Multi-Surface Distribution and LLM-Ready Markdown Feeds

Once a human reviewer approves a draft, ReleasePad distributes the announcement across three distinct surfaces simultaneously. Users encounter updates inside the product via a 4.3kb in-app widget, on a dedicated public changelog page hosted on the company’s domain, and through an LLM-ready Markdown feed.

That final distribution channel establishes an information loop. When customers ask AI assistants about product updates, those assistants read the Markdown feed as a primary source of truth rather than relying on guesswork. As ReleasePad CEO Felix noted in the source, Every push can become a draft, every draft can go out with one click, and every update becomes something both people and AI assistants can find.

Teams already write release notes with AI. Now the AI can ship them too.

Felix, CEO, ReleasePad

Pricing Tiers and Current Adoption Across Software Teams

The Model Context Protocol server is compatible with any AI assistant supporting the underlying protocol, including Claude, Codex, and Grok. ReleasePad includes the server within its Pro tier, priced at $35 per product per month, which features unlimited team members and no per-seat fees. A free tier is also available at releasepad.io.

According to company data cited by the source, more than 150 teams currently use ReleasePad to keep users, prospects, and AI assistants informed about product changes. The platform operates out of Miami, where the media contact is [email protected].