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

Superpowers: How I'm using coding agents in October 2025

blog.fsck.com

Post reply on HN
Print view View original

221–230 of 242 posts

Re: Superpowers: How I'm using coding agents in October 2025

#221

Post by a123b456c »

aydyn wrote: post #187
benrutter wrote: post #179

I'm so curious around what people's median experience is of AI coding tools. I've tried agents every now and then, recently for something very simple- add an option to request csb format in a data api. The results were, well, not good. . . I ended up undoing literally all changes because writing from scratch was a lot easier than trying to refactor the total mess it has made from what I'd have thought was a trivial f…

Think of this: whats the likelihood that what you are asking for would be found in some public github repo? If its high then you are good to go.

I think you're pointing in the right direction, but I would rephrase as,

what's the likelihood that the solution exists in the github repo in a way that the machine can recognize as relevant to your prompt?

If many versions of the solution exist, due to the problem's common occurrence, and if you can evaluate the LLM's output, then you're good to go.

Re: Superpowers: How I'm using coding agents in October 2025

#222

Post by kragen »

Aloisius wrote: post #178
kragen wrote: post #170

Earlier quoted context omitted.

That's just the copyright office of one country out of a couple hundred, the courts can overrule them, and legislation can change. However, I agree that currently in the US (or on code written in the US) copyright probably doesn't inhere in AI-written code.

The US constitution limits copyright to protection for authors and inventors. I'm skeptical that a simple law could extend protection to machine generated works without being ruled unconstitutional nor does there appear to be any significant government or public support for such a thing. And while yes, the US is just one country, but it does have a bit of an outsized software development industry. I also haven't hear…

The US already extends copyright to the output from compilers, on the flimsy basis that it is a "literary work", and enacted a sui generis 20-year "mask works" right for chip layouts, which are generally output from EDA tools. It's hard to predict what politics will do, except in the very general sense that policies that have no constituency will not be enacted.

Re: Superpowers: How I'm using coding agents in October 2025

#224

Post by mattmanser »

sothatsit wrote: post #203
benrutter wrote: post #179

I'm so curious around what people's median experience is of AI coding tools. I've tried agents every now and then, recently for something very simple- add an option to request csb format in a data api. The results were, well, not good. . . I ended up undoing literally all changes because writing from scratch was a lot easier than trying to refactor the total mess it has made from what I'd have thought was a trivial f…

Agent performance depends massively on the work you do. For example, I have found Claude Code and Codex to be tremendously helpful for my web development work. But my results for writing Zig are much worse. The gap in usefulness of agents between tasks is very big. The skill ceiling for using agents is also surprisingly high. Planning before coding, learning agent capabilities, environment setup, and context engineer…

I feel there's a third reason.

When I see a pro-AI person insisting that they are fully automated, I often scour their recent comments to find code or git repos they have shared. You find something every now and again.

My thinking is that I want to use this stuff, but don't find the agentic AI at all effective. I must be doing something wrong! So I should learn from the real world success of others.

A regular pattern is they say they're using vibe coding for complex problems. You check, and they're trivial features.

One egregious example was a basic randomizer to pick a string from a predetermined set, and save that value into an existing table to re-use later.

To me that's a trivial feature, a 15-30 minute task in a codebase I'm familiar with.

For this extremely AI bullish developer it was described as a major feature. The prompts were timestamped and it took them 1/2 day using coding agents.

They were sharing their .claude folder. It had 50 odd md files in it. I sampled a bunch of them and most of them boiled down to:

'You are an expert [dev/QA/architect/PM/tester]. Ultrathink. Be good'.

Worse, I looked at their linkedin, and on paper they looked experienced. Seeing their code, they were not.

There's a subset of the "fully automated" coders who are just bad. They are incapable of judging how bad AI code is. But vocally, and often aggressively, advocate for it.

Some are good, but I just can't replicate their success. And they're clearly also still hand-writing a lot of the code.

Re: Superpowers: How I'm using coding agents in October 2025

#225

Post by make3 »

danshapiro wrote: post #194
bcoates wrote: post #80

Earlier quoted context omitted.

Also the persuasion paper he links isn't at all about what he's talking about. That paper is about using persuasion prompts to overcome trained in "safety" refusals, not to improve prompt conformance.

Co-Author of the paper here. We don't know exactly why modern llms don't want to call you a jerk, or for that matter why persuasive techniques convince them otherwise. it's not a hard line like many of the guardrails. That said, I talked to Jesse about this, and I strongly suspect the same techniques will work for prompt conformance when the topic is something other than name calling.

isn't that just instruction fine tuning and rlhf inducing style & deference? why is that surprising

Re: Superpowers: How I'm using coding agents in October 2025

#226

Post by sothatsit »

mattmanser wrote: post #224
sothatsit wrote: post #203

Earlier quoted context omitted.

Agent performance depends massively on the work you do. For example, I have found Claude Code and Codex to be tremendously helpful for my web development work. But my results for writing Zig are much worse. The gap in usefulness of agents between tasks is very big. The skill ceiling for using agents is also surprisingly high. Planning before coding, learning agent capabilities, environment setup, and context engineer…

I feel there's a third reason. When I see a pro-AI person insisting that they are fully automated, I often scour their recent comments to find code or git repos they have shared. You find something every now and again. My thinking is that I want to use this stuff, but don't find the agentic AI at all effective. I must be doing something wrong! So I should learn from the real world success of others. A regular pattern…

Yeah, I definitely see this as well. These are the people with seven MCP servers, 5000-line AGENTS.md files, their own "memory systems" for the agents, and who try to hit their rate-limits on all their agents every 5 hours (regardless of whether or not they are actually getting useful work done). Having tried some of this stuff when I was trying to learn about agents, it almost always made their performance worse...

In web development, where I get the most out of agents, I am still only using them for implementing basic things. I will write anything even moderately complex, as agents often make the wrong assumptions somewhere. And then there's also manual work required to review and tidy up agent output. But there's just so much grunt work in web development from adding to a DB schema, writing a migration, adding the data to your model, exposing it in an API endpoint, and finally showing it on a page. None of that is complicated, so agents are pretty good at it.

Re: Superpowers: How I'm using coding agents in October 2025

#227

Post by mceachen »

no-name-here wrote: post #196
mceachen wrote: post #134

Earlier quoted context omitted.

"Refrain from including estimated task completion times." has been in my ~/.claude/CLAUDE.md for a while. It helps.

Do such instructions take up a tiny bit more attention/context from LLMs, and consequentially is it better to leave it off and just ignore such output?

I have to balance this with what I know about my reptile brain. It’s distracting to me when Claude declares that I’m “absolutely right!” or making a “brilliant insight,” so it’s worth it to me to spend the couple context tokens and tell them to avoid these cliches.

(The latest Claude has a `/context` command that’s great at measuring this stuff btw)

Re: Superpowers: How I'm using coding agents in October 2025

#228

Post by d4rkp4ttern »

iamjfu wrote: post #216

I am interested by this link: https://blog.fsck.com/blog/2025/superpowers/superpowers-demo... ``` Claude Code v2.0.13 Sonnet 4.5 (with 1M token context) Claude Max /Users/jesse/tmp/new-tool/.worktrees/todo-cli ``` How does this person have access to Sonnet 4.5 with 1m token context? I don't see this referenced anywhere when I search or when I ask Claude about it.

It’s a limited release beta feature not available to all. You can try to activate it by doing: /model sonnet[1m] And it accepts it but the at the next API call it may fail and say “this beta model is not available with your subscription”.

I haven’t gotten access yet.

One of the nice things about Codex (GPT-5) is the supposed 400k token context (although performance starts to deteriorate when you get to 80% context usage).

Re: Superpowers: How I'm using coding agents in October 2025

#229

Post by d4rkp4ttern »

philbo wrote: post #56
spprashant wrote: post #26

I am not ashamed to admit this whole agentic coding movement has moved beyond me. Not only do I have know everything about the code, data and domain, but now I need to understand this whole AI system which is a meta skill of its own. I fear I may never be able catch up till someone comes along and simplifies it for pleb consumption.

I think this and other recent posts here hugely overcomplicate matters. I notice none of them provides an A/B test for each item of complexity they introduce, there's just a handwavy "this has proved to work over time". I've found that a single CLAUDE.md does really well at guiding it how I want it to behave. For me that's making it take small steps and stop to ask me questions frequently, so it's more like we're pai…

Indeed Anthropic’s best practices suggest keeping the CLAUDE.md relatively small.

Re: Superpowers: How I'm using coding agents in October 2025

#230

Post by piperswe »

iamjfu wrote: post #216

I am interested by this link: https://blog.fsck.com/blog/2025/superpowers/superpowers-demo... ``` Claude Code v2.0.13 Sonnet 4.5 (with 1M token context) Claude Max /Users/jesse/tmp/new-tool/.worktrees/todo-cli ``` How does this person have access to Sonnet 4.5 with 1m token context? I don't see this referenced anywhere when I search or when I ask Claude about it.

OpenRouter shows Sonnet 4.5 as having a 1M context limit: https://openrouter.ai/anthropic/claude-sonnet-4.5
Post reply on HN