raiyanyahya

by Raiyan Yahya

Durable, fully-local project memory for Claude Code. Logs every session to .recall/history.md and condenses it into .recall/context.md with a local Python summarizer (TF-IDF + TextRank) — no API key, no external model, nothing leaves your machine. New sessions resume from the saved context.

by Raiyan Yahya

The world's most advanced learning system, inside Claude Code. Type /superlearn <topic> — it researches the live web, designs a page that fits the subject, and serves a local web app for deep learning: concept deep-dives, diagrams, typeset math, charts, runnable code, curated papers and lectures, a Miro-style whiteboard of the whole topic, focus mode, spaced repetition, and highlights and notes that Claude reads back — hot-reloading as you keep prompting the session.

by Raiyan Yahya

Claude does the work and replies with one word: done. Cuts summaries, preamble, and wrap-up chatter to save output tokens and usage.

by Raiyan Yahya

Convene a council of LLMs that independently propose, peer-review, rebut, vote, and synthesize a group-confirmed consensus answer — exposed as the ensemble_debate tool plus an /ensemble command.

by loop contributors

Write and run Loopfiles: agent loops with memory, verified done, a critic, and brakes.

by raiyanyahya

If-this-then-that rules for Claude Code: gate git pushes on tests and scans, protect files, auto-format after edits, and steer the agent — all from one readable rules file.