From deda7848c6b9635d61294420bd2c8502ea060ff6 Mon Sep 17 00:00:00 2001 From: Parfii-bot Date: Wed, 22 Apr 2026 02:58:05 +0800 Subject: [PATCH] docs(readme): move 'From the author' to opening, expand with transformer-error context --- README.md | 42 ++++++++++++++++++++++++------------------ 1 file changed, 24 insertions(+), 18 deletions(-) diff --git a/README.md b/README.md index 6ffed42..cdf7ae2 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,29 @@ # KeiSeiKit — Constructor-Pattern Agent Kit for Claude Code +## From the author + +Hello. + +Transformers are statistically wired to lie. It is not a bug, it is the core operation: they pick the next token by sampling a probability distribution conditioned on whatever context happens to fit the window. They cannot reliably hold long context, they will drag a hallucination picked up three thousand tokens ago into a confident final sentence, and they may deliver a brilliant insight right next to a fabricated citation. This is not fixable inside the model — it is mathematically baked in from the moment of tokenization onward. + +This kit is my humble attempt to build scaffolding *around* those errors. Not to fix the transformer, but to make it behave a little closer to my own working logic: catch the common failure modes before they reach a commit, give it external memory that survives session boundaries, give it a rhythm resembling human work (day sessions → overnight consolidation → morning report), and let parallel agents coordinate through a shared state instead of stepping on each other. + +I work across 4 to 8 parallel Claude terminals most days. The problems I hit are mundane: forgetting between sessions, repeating the same mistake on the third try, parallel agents clobbering each other's files, a hallucinated API name shipped to production. None of these are solvable by a better prompt. They are solvable by structure around the prompt. + +Most of what is here is well-established bricks (git-as-state, cron, TF-IDF, constructor-pattern composition). What may be new in the Claude Code context is the Constructor Pattern for agents (composable blocks, deterministic build, rebuild-on-block-edit hooks) and sleep-sync (using a git repo as the transport layer between sessions, with an Anthropic-cloud agent doing nightly REM-style consolidation). + +It is not a product. It helps me personally. If it resonates with you, let me know. If enough feedback comes in, there will be a next version — more primitives, more patterns against the "forgetful" transformer. But that needs input; without it, I just keep using this quietly myself. + +Forks and PRs welcome from everyone, not only from those who write code. If you hit a problem with Claude Code and have an idea for solving it, open an issue with the description. A well-formulated problem is already half the solution. + +Hope it is a small Kei for someone to make vibecoding better. + +Thanks. + +--- + +## What it is + KeiSeiKit is a comprehensive drop-in toolkit for [Claude Code](https://claude.com/claude-code). It ships a curated set of composable behavioral blocks, a Rust assembler that builds agent `.md` files from TOML manifests deterministically, nine pre-wired hooks (three of them dedicated to RULE 0.14 session self-audit), 35 portable skills (including an interactive `/new-agent` wizard, 10 hub-and-spoke pipelines, and the `/self-audit` retrospective skill), **9 Rust primitive crates**, 13 shell primitives, and 11 cross-tool bridge templates. Everything follows a Constructor Pattern: one file per concern, manifests as single source of truth, and the generated agent files are regenerated on every relevant edit. The kit is MIT-licensed and fully generic — install it on a fresh machine and you get a sane 12-agent fleet (implementers, critics, researchers, cost-guardians, and more — all namespaced under `kei-*` so they won't collide with your own same-named agents), a wizard for spinning up new project specialists, 10 pipeline skills that combine primitives end-to-end (`/compose-solution`, `/site-create`, `/schema-design`, `/observability-setup`, `/auth-setup`, `/api-design`, `/ci-scaffold`, `/test-matrix`, `/docs-scaffold`, `/new-project`, `/vm-provision`), and a build pipeline that keeps every agent derivable from its manifest. @@ -329,21 +353,3 @@ See `skills/compose-solution/SKILL.md` and its phase files (`phase-1-intake.md` ## License MIT. See `LICENSE` in this directory. - -## From the author - -Hello. - -This kit is not an attempt to revolutionize anything, nor is it a production-ready product. It grew out of how I work day-to-day — usually across 4 to 8 parallel Claude terminals. I kept hitting the same problems: forgetting between sessions, parallel agents stepping on each other, the same mistakes on repeat. - -So I tried a human-inspired approach. Give the transformer a rhythm closer to how people work: daytime sessions → overnight sleep that consolidates the day's traces → morning report. Plus a "sleep on it" queue — drop a hard question in the evening, find a thought-through answer in the morning. - -Most of what's here is well-established bricks (git-as-state, cron, TF-IDF). What may be new in the Claude Code context is the Constructor Pattern for agents (composable blocks, deterministic build, rebuild-on-block-edit hooks) and sleep-sync (using a git repo as the transport layer between sessions). - -It helps me personally. If it resonates with you, let me know. If enough feedback comes in, there will be a next version — more primitives, more patterns against the "forgetful" transformer. But that needs input — without it, I just keep using it quietly myself. - -Forks and PRs are welcome from everyone, not only from those who write code. If you have a problem working with Claude Code and an idea for solving it, open an issue with the description. A well-formulated problem is already half the solution. - -Hope it's a small Kei for someone to make vibecoding better. - -Thanks.