> Checkpoints are a new primitive that automatically captures agent context as first-class, versioned data in Git. When you commit code generated by an agent, Checkpoints capture the full session alongside the commit: the transcript, prompts, files touched, token usage, tool calls and more.

This thread is extremely negative - if you can't see the value in this, I don't know what to tell you.

Ignoring the VC economics and awful name, I won’t be as pessimistic as everyone. I see the vision.

That said, nobody knows what the AI future looks like. Entire’s entire thesis is a solution for something we don’t even know we need. It’s a massive bet and uphill battle. Traditionally, dev tool success stories come from grassroots projects of developers solving their own problems and not massive VC funded efforts that tell you what you need to do.

Either the models are good and this sort of platform gets swept away, or they aren’t, and this sort of platform gets swept away.

>CLI to tie agent context into Git on every push.

Is this the product? I don't want to jump on the detractor wagon, but I read the post and watched the video, and all I gathered is that it dumps the context into the commit. I already do this.

We went from having new JavaScript frameworks every week to having new AI frameworks every week. I'm thinking I should build a HN clone that filters out all posts about AI topics...

60 million SEED round? This is really a thing now?

I see the vision here. I think this is extremely needed.

The context preservation problem is genuinely painful - I've been using task.md files and CLAUDE.md conventions to maintain agent state across sessions, and it's duct tape at best. First-class "checkpoints" that capture reasoning alongside diffs is an appealing idea.

But I'm skeptical of building this as a separate platform rather than as tooling on top of git. The most useful AI dev workflow improvements I've seen (cursor rules, aider conventions, claude hooks) all succeeded precisely because they stayed close to existing tools. The moment you ask developers to switch their entire SDLC stack, adoption becomes the real engineering challenge - not the tech.

Curious whether the open source commitment means the checkpoint format itself will be an open spec that other tools can build on.

> The game has changed. The system is cracking.

Just say what your thing does. Or, better yet, show it to me in under 60 seconds.

Web sites are the new banner ads and headings like that are the new `<blink>`.

> Spec-driven development is becoming the primary driver of code generation.

This sounds like my current "phase" of AI coding. I have had so many project ideas for years that I can just spec out, everything I've thought about, all the little ideas and details, things I only had time to think about, never implement. I then feed it to Claude, and watch it meet my every specification, I can then test it, note any bugs, recompile and re-test. I can review the code, as you would a Junior you're mentoring, and have it rewrite it in a specific pattern.

Funnily enough, I love Beads, but did not like that it uses git hooks for the DB, and I can't tie tickets back to ticketing systems, so I've been building my own alternative, mine just syncs to and from github issues. I think this is probably overkill for whats been a solved thing: ticketing systems.

Agents can save their reasoning into markdown files, and commit those files to Git. Are "Checkpoints" just a marketing term for that, or there's more to it?

Christ, a $60m seed round.

The AI fatigue is real, and the cooling-off period is going to hurt. We’re deep into concept overload now. Every week it’s another tool (don’t get me started on Gas Town) confidently claiming to solve… something. “Faster development”, apparently.

Unless you’re already ideologically committed to this space, I don’t see how the average engineer has the energy or motivation to even understand these tools, never mind meaningfully compare them. That’s before you factor in that many of them actively remove the parts of engineering people enjoy, while piling on yet another layer of abstraction, configuration, and cognitive load.

I’m so tired of being told we’re in yet another “paradigm shift”. Tools like Codex can be useful in small doses, but the moment it turns into a sprawling ecosystem of prompts, agents, workflows, and magical thinking, it stops feeling like leverage and starts feeling like self-inflicted complexity.

I am not willing to share my sheepish prompts with my team. Sorry!

I don't understand how this is different from giving an agent access to github logs? The landing page is terrible at explaining what it does.I guess they are just storing context in git aswell?

So is this just a few context.md files that you tell the agent to update as you work and then push it when you are done???

Actually interesting, but how's that different from just putting your learning / decision context into the normal commit text (body) ? An LLM can search that too, and doesn't require a new cli tool.

EDIT: Or just keep a proper (technical) changelog.txt file in the repo. A lot of the "agentic/LLM engineering frameworks" boil down to best approaches and proper standards the industry should have been following decades ago.

Huh, the checkpoint primitive is something that I've been thinking about for a while, excited to see how it's implemented in the CLI. Git-compatible structures seem to be a pretty big pull whenever they're talking about context management.

I shall give the benefit of a doubt given they are "building in the open". I feel my current setup already does all this though, so I struggle to see the point

This guy was the ex-ceo of GitHub and can't bother to communicate his product in a single announcement post?

Ironically, I was shortly contracting on PoC similar to this for ex github cofunder around this time last year.

This is a good idea but I feel like you could get something similar by just adding an instruction for the agent to summarize the context for the commit into a .context/commit/<sha> file as a git hook.

I had a similar, admitted poorly thought out idea a few months back.

I wanted to more or less build Jira for agents and track the context there.

If I had to guess 60 million is just enough to build the POC out. I don't see how this can compete though, Open AI or Anthro could easily spin up a competitor internally.

"$60M Seed round"

I guess when you are Ex-Github CEO, it is that easy raising a $60M seed. I wonder what the record for a seed round is. This is crazy.

I miss the good old days:

https://news.ycombinator.com/item?id=338286

I built a skill for this: https://github.com/doubleuuser/rlm-workflow

The readme is a bit more to the point.

Checkpoints sounds like an interesting idea, and one I think we'll benefit from if they can make it useful.

I tried a similar(-ish) thing last year at https://github.com/imjasonh/cnotes (a Claude hook to write conversations to git notes) but ended up not getting much out of it. Making it integrated into the experience would have helped, I had a chrome extension to display it in the GitHub UI but even then just stopped using it eventually.

disclosure: i run a startup that will most likely be competitive in the future.

I welcome more innovation in the code forge space but if you’re looking for an oss alternative just for tracking agent sessions with your commits you should checkout agentblame

https://github.com/mesa-dot-dev/agentblame

There is also Git AI: https://github.com/git-ai-project/git-ai https://usegitai.com/

My first thought that it was made for companies which tie "AI usage" to performance evaluation.

Sorry for not contributing to the discussion (as per the guidelines), but is it just me or this blog post reads a lot like LLM-filled mumble jumble? Seems like I could trim half of the words there and nothing would be lost.

I bet it will down/unstable 3/4 of the month.

This feels a bit like when some Hubbers broke off to work on PlanetScale, except without the massively successful, proven-to-be-scalable open source tool to build off (Vitess).

If you're approaching this problem-space from the ground up, there are just so many fundamental problems to solve that it seems to me that no amount of money or quality of team can increase your likelihood of arriving at enough right answers to ensure success. Pulling off something like this vision in the current red-ocean market would require dozens of brilliant ideas and hundreds of correct bets.

What's the long-term or even short-term strategy to make money?

It's not like $60m in funding was given as charity.

Think of all of the habit tracker and to do list apps we'll be able to make now!

The founder has only forked repositories on GitHub that are sort of light web development related.

His use of bombastic language in this announcement suggests that he has never personally worked on serious software. The deterioration of GitHub under his tenure is not confidence inspiring either, but that of course may have been dictated by Nadella.

If you are very generous, this is just another GitHub competitor dressed up in AI B.S. in order to get funding.

Entire.io, the name is on point considering it asks for access to my entire GitHub account.

But seriously, $300M valuation for a CLI tool that adds some metadata to Git commits. I don't know what to say.

I’m manually checking in Agent.md for every commit to improve the context window usage. Is that now automated?

Really hope that unlike GitHub it'll be open source

Highly dubious of this.

I see zero reason for a person to care about the checkpoints.

And for agents, full sessions just needlessly fill context.

So not sure what is being solved by this.

Just have a data lake with annotated agent sessions and tool blobs (you should already be keeping this stuff for evals), then give your agent the ability to query it. No need for a special platform, or SaaS.

As for SDLC, you can do some good automations if you're very opinionated, but people have diverse tastes in the way they want to work, so it becomes a market selection thing.

The lack of explanation of what it is and does is a tell of what gullible audience they are seeking.

Tech marketing has become a lot like dating, no technical explanation and intellectual honesty, just word words words and unreasonable expectations.

People usually cannot be honest in their romantic affairs, and here it is the same. Nobody can state: we just want to be between you and whatever you want to accomplish, rent seeking forever!

Will they ever care to elaborate HOW things works and the rationale behind stating this provides any benefit whatsoever? Perhaps this is not intended for those type of humans that care about understanding and logic?

Its a shame Pierre shut down. Wish they could have made it work. Github but made by Linear would be a dream.

I'm interested to see if they will try to tackle the segregation of human vs AI code. The downside of agents is that they make too much changes to review, I prefer being able to track which changes I wrote or validated from the code the AI wrote.

For people trying to understand the product (so far), it seems that entire is essentially an implementation of the idea documented by http://agent-trace.dev.

Can someone please explain what is this?

I am already overloaded with information (generated by AI and humans) on my day to day job, why do I need this additional context, unless company I work for just wants to spend more money to store more slop?

How is it different than reversing it, given a PR -> generate prompt based on business context relevant to the repo or mentioned issues -> preserve it as part of PR description

I barely look at git commit history, why should I look for even higher cardinality data, in this case: WTF, are you doing, idiot, I said don't change the logic to make tests pass, I said properly write tests!

> Cursor's Composer 2.0

There is no Composer 2.0. There is Cursor 2.0 and Composer 1.5.

There have been so many GitHub CEOs I was excited to find out which one.

New agent framework / platform every week now. It's crazy how fast things move...just when you get comfortable with an AI flow something new comes out...

> ... to Cursor's Composer 2.0 and more, ...

I couldn't find any references of Composer 2.0 anywhere. When did that come out?

I don't see how we need a brand new paradigm just because LLMs evidently suck at sharing context in their Git commits. The rules for good commits still apply in The New Age. Git is still good enough, LLMs (i.e. their developer handlers) just need to leverage it.

Personally, I don't let LLMs commit directly. I git add -p and write my own commit messages -- with additional context where required -- because at the end of the day, I'm responsible for the code. If something's unclear or lacks context, it's my fault, not the robot's.

But I would like to see a better GitHub, so maybe they will end up there.

Clicks through to see what Tom or Chris started…

Oh, nevermind, it’s some MS dude.

Wait, since when is Dohmke out? I thought this was gonna be Nat.

$300kk valuation for git commits :) the bubble will pop at some point, I don’t know when, but boy will it be spectacular.

Not sure what it is or what it does.

Which CEO?

$60M seed to wrap git hooks in YAML config. The AI tooling bubble is just VCs subsidizing solutions looking for problems while developers want less complexity, not more.

I don't want agent context tied to git commits. I just want infinite scroll in Claude Code and ability to search and review all my past conversations!

I did test it and use it and trashed it because there is very little value, actually none for me. These problems are easily being solved in other ways whoever has any experience with these tools. Getting $60M round for this stuff is ridiculous.

so github ci/cd agents rebranded as a startup? same team different company.

Sounds very cringe

Looking at the CLI implementation. Why not build on top of jj?

I really hate this trend of naming companies using dictionary words just because they can afford to spend cash on the domain name instead of engineering. Render, fly, modal, entire and so on.

Grifters to the grift god

Really struggling to figure out what this is at a glance. Buried in the text is this line which I think is the tl;dr:

"As a result, every change can now be traced back not only to a diff, but to the reasoning that produced it."

This is a good idea, but I just don't see how you build an entire platform around this. This feels like a feature that should be added to GitHub. Something to see in the existing PR workflow. Why do I want to go to a separate developer platform to look at this information?

List of Github CEOs:

1. Tom Preston-Werner (Co-founder). 2008 – 2014 (Out for, eh... look it up)

2. Chris Wanstrath (Co-founder). 2014 – 2018

(2018: Acquisition by Microsoft: https://news.ycombinator.com/item?id=17227286)

3. Nat Friedman (Gnome/Ximian/Microsoft). 2018 – 2021

4. Thomas Dohmke (Founder of HockeyApp, some A/B testing thing, acquired by Microsoft in 2014). 2021 - 2025

There is no Github CEO now, it's just a team/org in Microsoft. (https://mrshu.github.io/github-statuses/)

I thought something got seriously wrong with Nat Friedman but fortunately it's another one.

Oh man I'm tired. This reminds me of the docker era. It's all moving fast. Everyone's raising money. And 24 months from now it's all consolidating. It's all a nice hype game when you raise the funding but the execution depends on people finding value in your products and tools. I would argue yes many of these things are useful but I'd also argue there's far too much overlap, too many unknowns and too many people trying to reinvent the whole process. And just like the container era I think we're going to see a real race to zero. Where most of the dev tools get open sourced and only a handful of product companies survive, if that. I want to wish everyone the best of luck because I myself have raised money and spent countless years building Dev tools. This is no easy task especially as the landscape is changing. I just think when you raise $60m and announce a cli. You're already dead, you just don't know it. I'm sorry.

[dead]

[dead]

[dead]

[dead]

[dead]

[flagged]