Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GitHub Agentic Workflows

github.github.io

Post reply on HN
Print view View original

81–90 of 149 posts

Re: GitHub Agentic Workflows

#81

Post by thulah »

dboreham wrote: post #58
qwertox wrote: post #51

I want to see where we're at in 2 years, because these last couple of months have been pretty chaotic (but in a good sense) in terms of agents doing things with other agents. I think this is the real wake-up-call, that these dumb and error-prone agents can do self-correcting teamwork, which they will hopefully do for us. Two years, then we'll know if and how this industry has completely been revolutionized. By then w…

Spoiler: this is how humans always worked. Even Einstein had his wife, Marcel Grossmann and Hilbert, among others.

And Stalin had Lysenko.

Re: GitHub Agentic Workflows

#82

Post by resquawk »

wiether wrote: post #55
woodruffw wrote: post #24

I find this confusing: I can see the value in having an LLM assist you in developing a CI/CD workflow, but why would you want one involved in any continuous degree with your CI/CD? Perhaps it’s not as bad as that given that there’s a “compilation” phase, but the value add there isn’t super clear either (why would I check in both the markdown and the generated workflow; should I always regenerate from the markdown whe…

I thought that it was to allow non-tech people to start making their own workflows/CI in a no/low-code way and compete against successful companies on this market. But the implementation is comically awful. Sure, you can "just write natural language" instructions and hope for the best. But they couldn't fully get away from their old demons and you still have to pay the YAML tax to set the necessary guardrails. I can'…

The egress firewall is active by default, see https://github.github.io/gh-aw/introduction/architecture/

We've fixed the example on the README to be a link, it should be clearer now what's going on.

Re: GitHub Agentic Workflows

#83

Post by resquawk »

woodruffw wrote: post #24

I find this confusing: I can see the value in having an LLM assist you in developing a CI/CD workflow, but why would you want one involved in any continuous degree with your CI/CD? Perhaps it’s not as bad as that given that there’s a “compilation” phase, but the value add there isn’t super clear either (why would I check in both the markdown and the generated workflow; should I always regenerate from the markdown whe…

We've added an FAQ on determinism here: https://github.github.io/gh-aw/reference/faq/#determinism

Re: GitHub Agentic Workflows

#84

Post by resquawk »

woodruffw wrote: post #44
zozbot234 wrote: post #27

Earlier quoted context omitted.

> I find this confusing: I can see the value in having an LLM assist you in developing a CI/CD workflow, but why would you want one involved in any continuous degree with your CI/CD? The sensible case for this is for delivering human-facing project documentation, not actual code. (E.g. ask the AI agent to write its own "code review" report after looking at recent commits.) It's implemented using CI/CD solutions under…

Sorry, maybe I phrased my original comment poorly: I agree there's value in that kind of "self" code-review or other agent-driven workflow; I'm less clear on how that value is produced (performantly, reliably, etc.) by the architecture described on the site.

For Continuous Documentation examples, see https://github.github.io/gh-aw/blog/2026-01-13-meet-the-work...

Re: GitHub Agentic Workflows

#85

Post by thulah »

This is insane stuff. Why are they pushing this nonsense on developers when the real money is in surveillance and web indexing?

People like Nadella must think that developers are the weakest link: Extreme tolerance for Rube Goldberg machines, no spine, no sense of self-protection.

I'll cancel my paid GitHub account though.

Re: GitHub Agentic Workflows

#86

Post by amluto »

> GitHub Agentic Workflows deliver this: repository automation, running the coding agents you know and love, in GitHub Actions, with strong guardrails and security-first design principles.

GitHub Actions is the last organization I would trust to recognize a security-first design principle.

Re: GitHub Agentic Workflows

#87

Post by threecheese »

spankalee wrote: post #66
onionisafruit wrote: post #53

I noticed this unusual line in go.mod and got curious why it is using replace for this (typically you would `go get github.com/Masterminds/semver/v3@v3.4.0` instead). replace github.com/Masterminds/semver/v3 => github.com/Masterminds/semver/v3 v3.4.0 I found this very questionable PR[0]. It appears to have been triggered by dependabot creating an issue for a version upgrade -- which is probably unnecessary to begin w…

This happens with all agents I've used and package.json files for npm. Instead of using `npm i foo` the agent string-edits package.json and hallucinates some version to install. Usually it's a kind of ok version, but it's not how I would like this to work. It's worse with renaming things in code. I've yet to see an agent be able to use refactoring tools (if they even exist in VS Code) instead of brute-forcing renames…

For the first, I think maintaining package-add instructions is table stakes, we need to be opinionated here. Agents are typically good at following them, if not you can fall over to a Makefile that does everything.

For the second, I totally agree. I continue to hope that agents will get better at refactoring, and I think using LSPs effectively would make this happen. Claude took dozens of minutes to perform a rename which Jetbrains would have executed perfectly in like five seconds. Its approach was to make a change, run the tests, do it again. Nuts.

Re: GitHub Agentic Workflows

#88

Post by threecheese »

siscia wrote: post #59

I am somehow close to what MSFT and GitHub are doing here, mostly because I believe it is a great idea, and I am experimenting on it myself. Especially on the angle of automatic/continuos improvement ( https://github.github.io/gh-aw/blog/2026-01-13-meet-the-work... ) Often code is seen as an artifact, that it is valuable by itself. This was an incomplete view before, and it is now a completely wrong view. What is val…

We are missing some building blocks IMO. We need a good abstraction for defining the invariants in the structure of a project and communicating them to an agent. Even if we had this, if a project doesn’t already consistently apply those patterns the agent can be confused or misapply something (or maybe it’s mad about “do as I say not as I do”).

I expend a lot of effort preparing instructions in order to steer agents in this way, it’s annoying actually. Think Deep Wiki-style enumeration of how things work, like C4 Diagrams for agents.

Re: GitHub Agentic Workflows

#89

Post by mbrumlow »

I think it is funny they all these companies are spending a ton and racing to have a AI story. It’s almost like none of the executives understand AI.

If you are changing your product for AI - you don’t understand AI. AI doesn’t need you to do this, and it doesn’t make you a AI company if you do.

AI companies like Anthropic, OpenAI, and maybe Google, simply will integrate at a more human leave and use the same tools humans used in the past, but do so at a higher speed, reliability.

All this effort wasted, as AI don’t need it, and your company is spending millions maybe billions to be an AI company that likely will be severely devalued as AI advances.

Re: GitHub Agentic Workflows

#90

Post by teaearlgraycold »

embedding-shape wrote: post #73
spankalee wrote: post #66

Earlier quoted context omitted.

This happens with all agents I've used and package.json files for npm. Instead of using `npm i foo` the agent string-edits package.json and hallucinates some version to install. Usually it's a kind of ok version, but it's not how I would like this to work. It's worse with renaming things in code. I've yet to see an agent be able to use refactoring tools (if they even exist in VS Code) instead of brute-forcing renames…

> This happens with all agents I've used and package.json files for npm. Instead of using `npm i foo` the agent string-edits package.json and hallucinates some version to install. When using codex, I usually have something like `Never add 3rd party libraries unless explicitly requested. When adding new libraries, use `cargo add $crate` without specifying the version, so we get the latest version.` and it seems to mak…

Eventually this specific issue will be RLHF’d out of existence. For now that should mostly solve the problem, but these models aren’t perfect at following instructions. Especially when you’re deep into the context window.
Post reply on HN