scribe

I've forked it locally, to be honest I haven't merged upstream in a while as I haven't seen any commits that I found relevant and would improve my usage, they seem to work on the web and desktop version which I don't use. The changes I've made locally are: - Added a discuss mode with almost on tools except read file, ask tool, web search only based no heuristics + being able to switch from discuss to plan mode. Experiments: - hashline: it doesn't bring that much benefit over the default with gpt-5.4. - tried scribe [0]: It seems worth it as it saves context space but in worst case scenarios it fails by reading the whole file, probably worth it but I would need to experiment more with it and probably rewrite some parts. The nice thing about opencode is that it uses sqlite and you can do experiments and then go through past conversation thr

The approaches used by Claude Code and Cursor are inefficient. It's possible to calculate a covering set for a piece of code and provide that to an agent directly via a tool, and it turns out that this can reduce context usage in SWE-bench style tasks by >90% over RAG and grep/read. If you're interested in learning more, https://github.com/sibyllinesoft/scribe

Just released a Code Quality leaderboard based on the static analysis research I've been doing. https://codehealth.sibylline.dev/ Now working on comprehensive benchmarks for another tool I built, https://github.com/sibyllinesoft/scribe. Results thus far showing it reducing agent token usage by ~80% in real world tasks, but I need to repeat to get variance. Hopefully I can get a writeup out soon.

Hi HN, I made this tool because I was tired of babysitting repomix to get the code I wanted to provide LLMs packed in large repositories. Scribe uses token budgeting and a variety of IR algorithms to "just work" on any repo with no configuration. Acknowledgements to repomix for a feature compatibility rubric and rendergit for the original HTML layout code.