npm i task-master-ai

our context graph came to life today - full decision lineage - self-healing, self-updating team state - graph expanding as the team works - knows why we did what did when we did it AS WE DO IT, across the whole team everyone’s building infrastructure to read organizational knowledge. we’re building where it gets created. it’s what happens when alignment generates context as exhaust instead of losing it to slack threads holy shit we are never going back

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

@EyalToledano @blader

@blader cool name not confusing at all but probably worth adjusting @TaskmasterAI

you know when you ask claude code to build something and you write up a nice plan, kick it off before you go to bed, and then wake up the next morning and found it decided to just stop 30 minutes in because it's lazy? yeah, i hate that too. you don't need ralph. you need a taskmaster. now, every time claude tries to stop, taskmaster will yell at it and go "hey. hey you! did you actually finish? no? then what the hell are you doing? keep going!" and poor claude will have to keep going. get it here: github.com/blader/taskmas…

@ycombinator @amiklas @usehamster

Cursor for Product Managers @amiklas AI is great at writing code, but product success depends on deciding what to build. We think there’s an opportunity for an AI-native system focused on helping teams figure out what to build, not just how to build it.

The way startups are built has shifted quickly. We're excited about a range of startup ideas for AI-native companies that can now be built faster, cheaper, and with more ambition than ever. ycombinator.com/rfs

@ukituki 🤷

We’re collectively discovering practices that have been available in @TaskmasterAI for months.

@d4m1n @EddyLeeKhane @m1tch37 maybe ‘tm loop’ already released as of 0.41 and we haven’t announced it yet :) github.com/eyaltoledano/c…

@EddyLeeKhane @m1tch37 @TaskmasterAI could be cool for TM to get loop support :P

before I go to bed, I will left Ralph rip on 116 tasks 🤠 will I wake up to a new personal website? this shouldn't be possible. stay tuned! for now, gn 😴

@iamjohnellison @ireaderinokun this man knows something

@ireaderinokun Use @TaskmasterAI or SpecKit with a really solid, well considered PRD for any build. Build task by task. Work to 70-80% context usage only then summarize work, compact / clear then continue.

Woke up decided today is the day I'm going to master Claude Code. Any key tips/resources you have?

@RalphEcom @davideasaf ralphmaster

@davideasaf @TaskmasterAI i’m thinking of doing a tm ralph which basically keeps looping on tm next and does the task with claude and keeps doing tm next until it can’t anymore

@TaskmasterAI, all this Ralph Wiggum stuff makes me laugh because it’s literally a while loop over taskmaster. I’ve been using Ralph without the name since last March! IMO, the latest models make a huge difference in the hype and chances of success.

@BardoDev @iamsahaj_xyz @opencode speak your wish and see it made true!

@iamsahaj_xyz I just tell my agent to use @TaskmasterAI to expand a PRD and plan that way. Then iterate on all the tasks lists until they are done. A tight taskmaster plug-in integration to @opencode would be sick!

@ryancarson @Z0rk153 you’re great

@Z0rk153 @TaskmasterAI TM is great

@davideasaf 🤷 we’ll release the harness soon as well :)

@iamjohnellison @BenjaminDEKR We’ve taken the lessons from Taskmaster and bringing them into @usehamster

@BenjaminDEKR Nice! I really like this. I have been using @TaskmasterAI and SpecKit but craving a more PM-style visual roadmap. What you’ve built seems next level! Are you productizing this?

I built an AI Factory that uses Claude Code agents to build apps while I sleep. This is an automated "assembly line" pipeline, all run by Claude Opus 4.5. Projects move through a Kanban-like system from Idea to Research, Architecture, Coding and Testing. Factory workers start with market research, searching the web and social media. Then they validate everything, checking app stores for competition and securing a domain name. (All automated by APIs and MCPs) They automatically create several rounds of app UI revisions. When I wake up, there are projects waiting for me to review: approve this design, give feedback, send that one back for improvement. Projects are actively coded and tested by the agents, while the entire process is tracked and logged (with "worker documents" that travel each step of the way, like a real factory.)

@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

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

@EyalToledano @linear @TaskmasterAI oh. can taskmaster be accessed from the web? for PM use

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

@joaopaulo_fs @H3roAi @TonyMcDow @donvito hello :)

@H3roAi @TonyMcDow @donvito Do you use some extra tool on top of claude code for orchestration? I do some as well, but mostly parallel agents to fix tests and etc

Should I give in to $100 Claude Max How’s Opus 4.5 limits in max