Claude Code is powerful, but juggling multiple projects in raw terminal tabs is a mess. Claudy wraps it in a native macOS app with proper multi-session management, auto account switching when you hit usage limits, Draft Commits to checkpoint and restore work mid-session, and a Marketplace to install Skills, MCPs, and Commands in one click.
I've always preferred running Claude Code directly in the terminal over IDE plugins. But once I started juggling multiple projects, it got messy fast — no clean way to manage sessions, projects, or accounts in one place.
That's what pushed me to build Claudy. It's essentially a native macOS wrapper that gives Claude Code proper multi-session and multi-project management, with auto account switching when you're close to hitting limits.
The Marketplace came later — I wanted an easy way to install and share Skills, MCPs, and Commands without manually copying files around. It's open-source, so anyone can submit their own.
Still early days. Would love to hear what features matter most to you!
The Marketplace for one-click MCP installation feels underrated. Manually installing and configuring MCPs is still a huge friction point for getting agents to do real-world work reliably. This honestly feels similar to what package managers were for developers in the early 2010s. Quick question: does the Marketplace handle versioning and dependency conflicts across MCPs, or are those managed manually for now?
The Draft Commits feature is the most underrated thing here. Claude Code sessions can go off the rails 20 minutes in and there's currently no clean way to checkpoint without manually stashing or branching. AI-labeled [Draft] commits that you can restore from and then squash into a clean final commit is exactly the right mental model — it maps how you actually think about agentic coding sessions, not how git works by default. Curious whether the AI labeling is happening locally or via an API call, and whether it's labeling based on the diff or the session conversation.
You know the pricing model is broken when you have to have multiple accounts to save money 😅 I guess you get away with that when you have the most successufl product in history.
the multi-account switch on rate limits is the part that would actually save me time — i currently keep two iterm windows open and swap api keys by hand. is account switching surfaced as a manual button or does it auto-fail-over when claude returns a 429?
About Claudy on Product Hunt
“A proper home for Claude Code — multi-session, multi-account”
Claudy launched on Product Hunt on May 13th, 2026 and earned 149 upvotes and 12 comments, placing #9 on the daily leaderboard. Claude Code is powerful, but juggling multiple projects in raw terminal tabs is a mess. Claudy wraps it in a native macOS app with proper multi-session management, auto account switching when you hit usage limits, Draft Commits to checkpoint and restore work mid-session, and a Marketplace to install Skills, MCPs, and Commands in one click.
Claudy was featured in Productivity (651.7k followers), Developer Tools (512.4k followers) and Vibe coding (452 followers) on Product Hunt. Together, these topics include over 202.2k products, making this a competitive space to launch in.
Who hunted Claudy?
Claudy was hunted by Mark G. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how Claudy stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Mark G