neuron-explorer
Steering-first neuron explorer for Qwen3.5-4B: push every MLP neuron, keep every rollout, browse the results
An entry point for developers new to agentic coding: how a rule you write once keeps holding in later sessions, checked while the agent builds rather than in a review you do after.
Developer toolsTypeScript · JavaScriptGitHub @krzysztofdudekOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
Say it once. Rules your AI coding agent still follows three sessions later, checked before it moves on. Free local checks, keyless CI.
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
Say it once.
Write a rule and it holds in every session after that, without you repeating yourself. Before the agent edits a file it gets only the rules that touch that file, not all two hundred. After the edit they are checked, and a violation comes back as an error the agent has to fix before it moves on. The same checks re-run in CI for free, with no API key.
You probably don't feel this problem, and that is the interesting partRead the full README ↗The full README is unavailable right now. Read it at the original link. The saved excerpt is shown below.
Say it once.
Write a rule and it holds in every session after that, without you repeating yourself. Before the agent edits a file it gets only the rules that touch that file, not all two hundred. After the edit they are checked, and a violation comes back as an error the agent has to fix before it moves on. The same checks re-run in CI for free, with no API key.
You probably don't feel this problem, and that is the interesting partRead the full README ↗