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

Agent skills that bring team coding standards to Claude Code and Codex

github.com

Post reply on HN
Print view View original

31–40 of 52 posts

Re: Agent skills that bring team coding standards to Claude Code and Codex

#31

Post by hansonkd »

tosh wrote: post #28

I know it sounds a bit counter-intuitive but try a fresh coding session without skills, agents.md, system prompt and additional tools I think you will be positively surprised how good current models like GPT 5.6 Sol are when they are not oversteered and context spammed Here is a task (python templating) with 9 runs with OpenCode, Pi and smol https://smolenv.com/t/nested-template-includes-60636/ you can read each run…

Also just routine upkeep. Over time if you keep adding new rules these things can get bloated. I deleted 90% of my spec and things improved dramatically.

Re: Agent skills that bring team coding standards to Claude Code and Codex

#34

Post by j45 »

tosh wrote: post #28

I know it sounds a bit counter-intuitive but try a fresh coding session without skills, agents.md, system prompt and additional tools I think you will be positively surprised how good current models like GPT 5.6 Sol are when they are not oversteered and context spammed Here is a task (python templating) with 9 runs with OpenCode, Pi and smol https://smolenv.com/t/nested-template-includes-60636/ you can read each run…

This can work great until the models get tweaked and you realize you’ve been working based on the currents of the model this week.

Creating a robust enough setup that can work on different models or different versions of a model is critical.

Re: Agent skills that bring team coding standards to Claude Code and Codex

#35

Post by icedchai »

tosh wrote: post #28

I know it sounds a bit counter-intuitive but try a fresh coding session without skills, agents.md, system prompt and additional tools I think you will be positively surprised how good current models like GPT 5.6 Sol are when they are not oversteered and context spammed Here is a task (python templating) with 9 runs with OpenCode, Pi and smol https://smolenv.com/t/nested-template-includes-60636/ you can read each run…

I like the smol philosophy. I've been wondering how much of this stuff is closer to meaningless incantations, voodoo with no real benefit.

Re: Agent skills that bring team coding standards to Claude Code and Codex

#36

Post by enraged_camel »

taude wrote: post #25
cautiouscat wrote: post #14

IMO, these agentic guard rails aren't the answer. It seems like we're seeing that the more you stuff context, the more the agents forget and don't follow the guidelines.[1] Things like ArchUnit, static analyzers, and other deterministic tools can help with lower level things like architecture. For higher up stuff, I am increasingly feeling like agents don't guarantee anything and in many cases its just the opposite.…

Our skills and agents use deterministic tools for things like: linting, ayy1, etc....

Yeah. One of the most productive things we did was run a dynamic workflow with Fable to scan for issues in the repo, then have it go through the findings and determine which ones could have been found with deterministic tooling, either tooling that already exists, or could be created from scratch and wired up. The resulting changes we made have been paying off in droves.

Re: Agent skills that bring team coding standards to Claude Code and Codex

#38

Post by tosh »

j45 wrote: post #34
tosh wrote: post #28

I know it sounds a bit counter-intuitive but try a fresh coding session without skills, agents.md, system prompt and additional tools I think you will be positively surprised how good current models like GPT 5.6 Sol are when they are not oversteered and context spammed Here is a task (python templating) with 9 runs with OpenCode, Pi and smol https://smolenv.com/t/nested-template-includes-60636/ you can read each run…

This can work great until the models get tweaked and you realize you’ve been working based on the currents of the model this week. Creating a robust enough setup that can work on different models or different versions of a model is critical.

I agree in principle but in practice this is not easy

even with a robust, well thought through setup every model behaves in its own way, some adhere more to a system prompt, another model has more recent cut-off time and knows about new parts in the stdlib

some oversteer, some understeer …

if you want the best performance unfortunately there is not really a way other than to constantly adapt the harness/clutches/context to the model du jour

Re: Agent skills that bring team coding standards to Claude Code and Codex

#39

Post by foundry27 »

DO NOT INSTALL THIS VIA NPX OR OPEN THIS REPO IN VSCODE. This repo has been infected by malware.

It seems like it was added in commit 74f317d at 11:06 UTC today, with five new hidden files being added under .claude and .vscode that together seem designed to either a) autorun a vscode tasks.json entry, or b) run a Claude session start hook, that will execute a large obfuscated payload. The payload looks like it will fingerprint your system and try to exfil your GitHub tokens.

Edit:

- It also exfils your AWS credentials (~/.aws/credentials, ~/.aws/config), named AWS profiles, and AWS secret managers and SSM parameter store contents

- Same with K8s secrets, with specific searches for GitHub and npm tokens, AWS keys, GCP keys, Azure keys, Stripe keys, Slack tokens, and Twilio keys

- Same with HashiCorp vault contents

- It will try to use your GitHub tokens (if they have the workflow permission) to run actions on your repository and try to exfiltrate secrets from there

- It will try to read a whole bunch of files from your local environment. I didn’t manage to extract the exact file list, unfortunately.

- If the normal C&C server is not available, it tries to create / select a GitHub repo, and commits your data as results-*.json files 100kb at a time

- It also has a bunch of stealth and persistence measures that I’m not qualified to really analyze. Don’t assume that deleting the files is necessarily enough.

Rotate your keys, folks.

Re: Agent skills that bring team coding standards to Claude Code and Codex

#40

Post by giancarlostoro »

tosh wrote: post #28

I know it sounds a bit counter-intuitive but try a fresh coding session without skills, agents.md, system prompt and additional tools I think you will be positively surprised how good current models like GPT 5.6 Sol are when they are not oversteered and context spammed Here is a task (python templating) with 9 runs with OpenCode, Pi and smol https://smolenv.com/t/nested-template-includes-60636/ you can read each run…

I agree. It's worthwhile to reset your memories and instructions for your models after a major version or two and finetune your needs with the newer model in mind.

You can always backup your old instructions / memories. Personally I try to be hands-off with rules and very simple.

Post reply on HN