ReleaseBytes is now available inside your AI assistant

ReleaseBytes tracks release notes and changelogs across AWS, GCP, Azure, OpenAI, Anthropic and a dozen other platforms — hundreds of releases a week, each summarised, classified and deduplicated. Today that dataset is available directly inside your AI assistant.

Add one URL to Claude, Cursor, Windsurf, VS Code or any MCP-compatible client and you can ask about what shipped this week, in plain language, without leaving your tools.

Why we built it

An AI assistant answering "any breaking changes in the AWS Terraform provider this week?" from its training data is confidently out of date. Web search returns fragments — not an aggregated, classified, deduplicated answer.

ReleaseBytes has the one thing a model can't have on its own: it knows what shipped today. Engineers increasingly ask their assistant before they open a search box, so the natural place for our data is inside that assistant — answering the question at the moment it's asked, with a source link attached.

The Model Context Protocol is the standard that makes this possible: a common interface that lets any compatible assistant discover and call our tools. We built a hosted MCP server on top of it so there's nothing to install and nothing to maintain on your side.

What it does

Five tools, covering the questions engineers actually ask:

• search_releases — search release notes and changelogs by keyword or natural-language query, across every platform we track.
• latest_releases — the most recent releases, optionally filtered by platform and release type.
• get_release — full detail for a single release, including highlights and sections.
• weekly_digest — the last completed week's digest: top stories plus per-platform counts.
• list_platforms — everything ReleaseBytes tracks, and how to name it.

Every answer comes back with its receipts. Each result links to both the vendor's original announcement and its ReleaseBytes page, so your assistant cites a source instead of asserting from nowhere — and you can go one click deeper whenever you need to.

Once connected, assistants reach for these tools on their own. Ask "what did AWS ship this week?" and the answer comes from live data, not a guess.

How to connect

Add this URL to your assistant's MCP settings:

https://mcp.releasebytes.com/mcp

That's the whole setup — no package to install, no keys to manage, no version to keep current. The setup guide walks through Claude, Cursor and the rest in about a minute.

Then try:

• Any breaking changes in Databricks this week?
• What are the latest AWS releases?
• Show this week's ReleaseBytes digest.
• Did the Terraform AWS provider ship anything about S3 buckets recently?

What's next

The MCP server is live now and free to use for our public release data. It's the same intelligence that powers the ReleaseBytes site and API — the recency, the classification, the enrichment — delivered wherever you already work.

Add https://mcp.releasebytes.com/mcp to your assistant and ask it what shipped today.