dsifry
dsifry on Hacker News
I have been helping people get onboarded with Claude Code and the orchestrator I wrote called Metaswarm [1) and the response has been way beyond my expectations.
But don't take my word for it, try it out for yourself, it is MIT licensed, and you can create new projects with it or add it to an existing project.
[1] https://github.com/dsifry/metaswarm
Posted by
dsifry on Tuesday February 03, 2026 @10:59AM
from the show-hn dept.
https://dsifry.github.io/metaswarm/
A few weeks ago I posted about GoodToGo https://news.ycombinator.com/item?id=46656759 - a tool that gives AI agents a deterministic answer to "is this PR ready to merge?" Several people asked about the larger orchestration system I mentioned. This is that system.
I got tired of being a project manager for Claude Code. It writes code fine, but shipping production code is seven or eight jobs —...
Posted by
dsifry on Monday February 02, 2026 @08:18PM
from the show-hn dept.
https://github.com/dsifry/metaswarm
A few weeks ago I posted about GoodToGo https://news.ycombinator.com/item?id=46656759 - a tool that gives AI agents a deterministic answer to "is this PR ready to merge?" Several people asked about the larger orchestration system I mentioned. This is that system.I got tired of being a project manager for Claude Code. It writes code fine, but shipping production code is seven or eight jobs —...
Some github comments are marked as sctionable, some have threads and suggestions, some are suggestions or are nitpicks. This provides you with a deterministic, reliable red/green approach that you cn use to enforce your policy. Give it a try and you will see how it is much more reliable than using a nondeterministic agent, especially for complex reviews!
Yeah, it is about making sure that EVERY actionable PR comment gets addressed - whwther by fixing, resolving, creating a new issue, commenting that it is a will not fix, or blocking for human review - and then giving you a clear deterministic check you can do to reliably enforce your policy.
Just to be clear - the hook is deterministic, but the subagent running with an mcp server loaded is not - and for medium/large PRs, it can run out of context window or just forget what it is trying to do and get lazy and say 'Everything is good, ready to merge!' when in fact tests are failing or there are still unaddressed PR comments.
I found that it has improved overall code quality significantly, at the cost of somewhat slower velocity. But it has meant fewer interruptions where the ai is just waiting for me, or saying "Everything is ready!" only to find that ci/cd failed or there were clearly existing comments/issues.
Lol, I thought it did a reasonably good job, but to each their own - this was the difference between releasing the project so others could use it with decent documentation, or not releasing and just using it internally. :)
No, it just prepares the PR - it doesn't automatically merge. That would be very dangerous, imho!
That's EXACTLY right. Ready to merge is an important gate, but it is very stupid to just merge everything without further checks/testing by a human!
Right! It doesn't assume that all comments are actionable, or need to be worked on. However, if you allow anyone to comment on your PRs, it could be a malicious vector. So don't let anyone review PRs on projects that you care about!!!
That's why I intentionally don't have this hooked into an ingest flow - you still get control over what issues/stories you want the agent swarm to work on... Just now, I can know that the code that was written has been reviewed and all comments have been fully addressed!
Yup! Software engineers aren't going to be out of work anytime soon, but I'm acting more like a CTO or VPE with a team of agents now, rather than just a single dev with a smart intern.
Sure, but that mcp still missed actionable comments that are marked as Out of Scope or Outside the PR - and this doesn't require having the context window loss of having another mcp instantiated, either. Anyway, give gtg a competitive look against the mcp - you should be able to see the difference
I did! The issue however, is having a clear, deterministic method of defining when the code review was 'done'. So the hooks can fire off subagents, but they are non-deterministic and often miss vital code review comments - especially ones that are marked in an inline comment, or are marked as 'Out of PR Scope' or 'Out of range of the file' - which are often the MOST important comments to address!
So gtg builds all of that in and deterministically determines whether or not there are any actionable comments, and thus you can block the agent from moving forward until all actionable comments are thoroughly reviewed, acted upon or acknowledged, at which point it will change state and allow the PR to be merged.
Posted by
dsifry on Saturday January 17, 2026 @04:55AM
from the show-hn dept.
https://dsifry.github.io/goodtogo/
I've been using Claude Code heavily, and kept hitting the same issue: the agent would push changes, respond to reviews, wait for CI... but never really know when it was done.It would poll CI in loops. Miss actionable comments buried among 15 CodeRabbit suggestions. Or declare victory while threads were still unresolved.The core problem: no deterministic way for an agent to know a PR is ready to...
Just used the service and had a great chat with Chap, one of the founders. There's something qualitatively different when you're talking to someone using voice - it makes it easier to articulate some of the bigger questions that aren't just syntax or "help me debug this" type of questions. Anyway, I think this is a great idea, and love that these guys are making it happen. Kudos.
Of course, that's completely true - and all the more reason to follow your love and passion if you happen to be privileged by luck of birth, family, and genetics. I think that this is true of many HN readers and certainly describes me. I'm incredibly grateful for being born into that lucky circumstance.
And I don't mean to put down people who can't do that because they have to figure out how to stay alive in a war torn country, or figure out how to keep your family from starving.
Thanks for your excellent point.
Just do what you love. Period. I've had times in my life when that was programming - and it got me a long way, and I loved doing it while I did it. Then I felt the challenge of learning how to build a business. So I made a choice, and stepped away from coding, and started pursuing my passion of building businesses. There's nothing wrong with your passions changing - think of it as opportunities to learn new skills.
One of the funny (paradoxical?) results of this is that I've had my eyes opened on things that I'd love to code. So, I've gotten passionate about programming again. Ain't life funny?
You don't need to apologize for being honest and self-examining - you only need to apologize if you aren't following your love and passion.
That's gotta be one of the craziest post titles I've read on HN lately ;-)
Technorati was a side project for me.
Posted by
dsifry on Thursday December 10, 2009 @06:19PM
from the ask-hn dept.
I'm looking to get a large monitor to use as a second monitor for coding projects. After doing some research, I've been blown away by how 23"+ monitors have come down in price. Ideally, the monitor should be large, responsive, and sharp, and something that doesn't need a dual-head video card to drive it. It should be reliable and have a decent warranty, so I'm leaning towards name brands, but...
The beta message is back in there. Thanks!
Great feedback, you make some very good points. This is, after all, what a beta is all about - to get out there, learn as much as we can, and iterate iterate iterate.
Thanks for being brutally honest, this is just what we want to learn!
Dave
Good point. We actually had the beta label in there, but somehow it got dropped out when we did our final QA. We'll get it back in there.