Hamster

@tomismeta @EyalToledano Keep shipping Tom!

@EyalToledano @usehamster Thanks so much, I'll check out Hamster. This is my first shot at building anything publicly. Zero ego here, just trying to exercise some unused muscles around an idea I like. Any feedback is a gift.

You used to wait for code. Now you wait for decisions.

every time you start a new claude code session you spend the first 10 minutes re-explaining your project "here's the stack. here's what we're building. here's what was decided last week. here's the file structure. DON'T TOUCH THIS PART WE JUST REBUILT IT" but it's not just your own context. your teammates shipped 3 new things since yesterday, two architectural decisions changed and a dependency got swapped out. you need to capture all of that too, or claude is building from a stale snapshot you won't remember everything. so claude compiles context from scratch, misses things, and pulls in a direction that might completely miss the point. now multiply that by every dev on your team, every session, all day literal coordinated drift what we do instead: the .hamster folder syncs decisions, plans, and context to the filesystem automatically. claude reads it and just knows. no re-explaining. no stale snapshots. everyone's always got a fresh .hamster folder which updates in real-time as the rest of the team (or you) ship decisions, tasks, briefs, methods, skills, etc try it. you'll love it → @usehamster

everyone on your team maintaining their own CLAUDE.md is a symptom, not a solution. the root cause is that shared context has nowhere to live and evolve as your team makes decisions. you can't fix it with naming conventions or a better folder structure. decisions need a home. not a static file someone remembers to update. a workspace that accumulates as the team works. when a decision changes, it changes once, and every agent on the team pulls from that. there's no need to keep the markdown files up to date or copy-paste anything anywhere. claude just knows. your PRDs, your tasks, your skill files, your CLAUDE.md are all projections of the same underlying decisions. if the decisions live somewhere real, everything else stays aligned. if they stay up to date automatically, the rate of progress can explode because the synchronization overhead disappears. that's what we've been building into hamster. decisions get a home. blueprint holds your product's current state. methods encode how agents should work. the projections stay aligned because the source of truth is real. dm if you want to try it GA reeealllly soon!

@nanomader @michaelkutsche *squeals*

@michaelkutsche @usehamster

Playing around with Google's Genie3. My daughter had this idea of creating a hamster sim. The result is mind blowing. I've been working with 3d software since the mid 90s (before becoming a character designer). It is unbelievable what is possible now.

@ninan_phillip @mntruell @TaskmasterAI New onboarding is in, everything will click when we ship Blueprint (ongoing)

@usehamster @mntruell @TaskmasterAI I got an invite for hamster a while back. Onboarding was a little confusing. I wasn’t exactly sure what I was supposed to be doing. Might have to give it another go. I really wanted taskmaster to work, but it didn’t have a great fit for our enterprise team.

How would you like to hack or extend Cursor? What does a perfectly extensible coding agent look like to you?

@ninan_phillip @mntruell @TaskmasterAI We’re building this. :)

@mntruell I think there should be ways to document PRDs, organize them, break them down, then have sub agents work on them. Kind of like Conductor and @TaskmasterAI meshed together. But his way you wan plan your entire work, break it down down, and have it worked at on a systematic way.

@EyalToledano @VizierPrime

@VizierPrime I agree - which is why the environment for where those decisions are made is up for innovation. It’s also the base primitive for how we’re building (and using) @usehamster - new briefs and PRDs are collaboratively built and alignment happens at that layer, not post-implementation

@EyalToledano @EddyLeeKhane @usehamster haha yeah i am very surprised to hear you’re using mean vc! that’s not what it’s for - but curious what aspect of it got you to use it for this. task master is built for this

@TaskmasterAI @donvito @EyalToledano @linear

@donvito @EyalToledano @linear for that you’ll want to use @usehamster, which is the web/multiplayer version of taskmaster. Built specifically for product teams

This is sick 🤯🤯🤯 You can monitor Claude Code tasks using a project management app Install the @linear MCP then ask Claude Code to make a plan first then save tasks to Linear. Ask it to update once each task is finished $ claude mcp add --transport sse linear-server mcp.linear.app/sse

we've shipped huge features in taskmaster v0.37 🚀 → real-time PRDs, team plans, synced tasks 🐹 → GPT-5, @lmstudio, @Zai_org GLM → @claudeai plugin + @aisdk v5 → smaller MCP tool footprint → redesigned init ux → enterprise proxy + more follow + retweet + bookmark 👀👇

we’ve built the core of @usehamster around multiplayer chat to power real-time collaboration between product + eng this leads to consolidated thinking, refinement, and team alignment for PRDs which produce flawless production-ready plans delivered in minutes with taskmaster

@nizzyabi shudder

do any newer startups use jira?

one of the big benefits of pm’s and designers using AI for prototypes is that code is becoming context at the discovery stage we can vibe prototypes to express ideas and interactions, get aligned and use the prototypes as context for the final delivery your vibed stuff doesn’t necessarily have to reach production. it can be used to express what you’re trying to do our approach to design @usehamster is very similar: designer designs lofi on figma, quickly brings it into claude code or cursor, prototypes it so the team can play with it, push it to a branch and then use that specific branch as context when drafting or refining new PRDs

@cobaltdigital33 @EyalToledano @levie we are hiring! tryhamster.com/careers

@EyalToledano @levie @usehamster Interesting product. Would love to contribute if possible

A core AI agent product management principle is just figuring out what a very smart person -without any initial context whatsoever- would need to perform the task successfully. The whole game is just doing everything possible to get just the right information into the context window to ensure that the agent gets access to the most relevant data and tools to execute. Every time we’re trying to figure out why something works or doesn’t work about an agent, usually it just boils down to the fact that a human would need totally different or meaningfully more context to execute the same action. Usually then the problem lies somewhere in the agent’s use of tools (like search), or not giving the agent enough data to work with, or sometimes giving it too much, or not explaining the task or objective properly, and so on. The great thing is that every one of these issues is tractable. The models will just keep getting better at every one of these issues. And you can always throw more compute at the problem in whatever form that is (more reasoning, more planning, more data retrieved, etc.) - it’s just a matter of cost/speed tradeoffs. Very interesting new space to be building for.

@vinogradovk thanks for the shoutout!

Over five years ago, I started publishing the top trending open-source startups every quarter — the ROSS Index. It has become a handy tool for 60K+ developers, founders, and other OSS fans who have visited it to date. Today, I'm thrilled to make it even more useful, as we are open-sourcing all its data for more than 20 quarters on GitHub, in a convenient format. Explore our repo and star it to be notified about new index releases. And, of course, now is also the time to reveal the fresh top 20 startups by GitHub stars growth for Q3 2025 — one can find more details on them in the repo too!

@pablosanzo @EyalToledano @TaskmasterAI yes, it’s nearly ready for early access. :)

@EyalToledano @TaskmasterAI Any product for that?

glad to see plan/spec modes taking off been parsing PRDs with @taskmasterai since march and found that: the more you focus on PRDs, the quicker you’ll realize the bottleneck isn’t describing the work itself we rarely write PRDs in a vacuum the challenge is team alignment