llmlb: Instructions file for Claude Code

CLAUDE.md

llmlb CLAUDE.md is an instructions file for Claude Code from akiojin/llmlb. It costs 10,946 tokens per session, scanned A, original, MIT.

A repository guidance file for Claude Code, an AI coding assistant. It explains the project layout, important terms, working rules, tests, and where to find more information.

In plain words
What is it for?
Use it to guide coding sessions in the llmlb repository, including setup, quality checks, test requirements, and project conventions.
Why use it?
It gives the assistant project-specific instructions instead of making it rediscover how the codebase is organised and how work should be done.

Instructions file for Claude Code

Written for Claude Code: PreToolUse hook event. Also seen: reads .claude/ paths; mentions .claude-plugin; mentions CLAUDE.md.

This is akiojin/llmlb's own configuration. It tells Claude Code how to work on llmlb itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything llmlb configures →

Reuse

Borrowing it

Nothing to install: this file belongs to akiojin/llmlb. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/akiojin/llmlb/develop/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/akiojin/llmlb

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for llmlb CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/akiojin/llmlb/claude-md/github.svg)](https://agentmods.dev/instructions/akiojin/llmlb/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/akiojin/llmlb/claude-md"><img src="https://agentmods.dev/badge/instructions/akiojin/llmlb/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for llmlb CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/akiojin/llmlb/claude-md"><img src="https://agentmods.dev/badge/instructions/akiojin/llmlb/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 10,946 This file is loaded in full into every session.
When invoked 10,946 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.10946 $0.10946
Opus 5.5 $0.04378 $0.04378
Sonnet 5.5 $0.02189 $0.02189
Haiku 4.5 $0.01095 $0.01095

Measured 8d ago against content hash b7c72758db83, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

llmlb CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 636 lines

How it starts

The opening of the file, as written. The whole thing — 636 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

このファイルは、このリポジトリでコードを扱う際のガイダンスを提供します。 短く「何を・どこを見るか」を示し、詳細は既存ドキュメントに段階的に委譲します (参考: Writing a good CLAUDE.md)。

まず読む 90秒版

  • 何を作る: Rust製ロードバランサー(llmlb/)+ llama.cppベースのC++推論エンジン(xLLM: https://github.com/akiojin/xLLM)。Ollamaは一切使わない/復活させない。
  • どこを見る: README.md(全体像)→ DEVELOPMENT.md(セットアップ)→ GitHub Issue(gwt-specラベル、要件とタスク)。
  • 守る: ブランチ/worktree作成・切替禁止、作業ディレクトリ移動禁止、必ずローカルで全テスト実行。
  • HFカタログ利用時はHF_TOKEN(任意)と必要に応じHF_BASE_URLを環境にセットしておく。
  • まず実行: make quality-checks(CI 相当の検証集合は Makefile の quality-checks ターゲットが定義する)。
  • 迷ったら: memory/constitution.mdとこのファイル後半の詳細ルールを再確認。
  • セッション開始時: memory/lessons.md を確認し、過去の修正パターンを踏まえて作業する。
  • 回答は日本語で行う。

ディレクトリ構成

llmlb/
├── llmlb/          # Rust製ロードバランサー(APIサーバー・管理UI・共通型定義)
├── xLLM (external)  # <https://github.com/akiojin/xLLM>
├── scripts/         # プロジェクトスクリプト(checks等)
├── memory/          # プロジェクト憲章・メモリファイル
├── docs/            # ドキュメント
├── .claude-plugin/   # Claude Codeプラグイン定義(skills含む)
├── poc/             # 概念実証・実験コード
└── vendor/          # サードパーティ依存(サブモジュール)

用語定義

用語 説明
llmlb(ロードバランサー) Rust製のAPIゲートウェイ。OpenAI互換APIを提供し、リクエストを適切なエンドポイントに振り分ける。ダッシュボード(管理UI)も内蔵。
エンドポイント llmlbが管理する推論サービスの接続先。xLLM、Ollama、vLLM、その他OpenAI互換APIなど多様なバックエンドを統一的に扱う。
xLLM 本プロジェクト独自のC++製推論エンジン。llama.cpp/whisper.cpp/stable-diffusion.cppなどを統合し、GPUを活用したローカル推論を提供。vLLM/Ollamaと同列のエンドポイントタイプとして扱う。

システムアーキテクチャ

ロードバランサー(llmlb/ - Rust製)

リクエストの受付・振り分け・管理を担当するコンポーネント。

  • OpenAI互換APIの提供: /v1/chat/completions, /v1/embeddings等のエンドポイント
  • 認証: JWT認証(ダッシュボード)、APIキー認証(API)
  • エンドポイント管理: 複数エンドポイントの登録・ヘルスチェック・負荷分散
  • モデルカタログ: HuggingFaceカタログとの連携、モデルメタデータ管理
  • リクエストルーティング: 適切なエンドポイントへのリクエスト転送
  • ダッシュボード: 管理UI(SPA)の提供

xLLM(external repo - C++製、llama.cppベース)

実際のLLM推論を担当するコンポーネント。vLLM/Ollamaと同列のエンドポイントタイプ。

  • モデルロード: GGUFファイルのロード・VRAM管理
  • 推論実行: テキスト生成、埋め込み計算
  • GPU管理: VRAM監視、モデルのロード/アンロード
  • ストリーミング: トークン単位のレスポンス

Read the full file on GitHub · 636 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 636 lines · 10,946 tokens per session scan A b7c72758db83

Subscribe to this mod's changes

llmlb CLAUDE.md is an instructions file published in the GitHub repository akiojin/llmlb (10 stars, last pushed 2d ago), licensed MIT. It adds 10,946 tokens to every session, about $0.0438 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-21.

Related

Other instructions, from other repositories

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 9,715 tokens

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,461 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).

microsoft/vscode · 5,262 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,680 tokens

vscode copilot-instructions.md

Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).

microsoft/vscode · 2,731 tokens

deepseek-harness AGENTS.md

AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.

deepseek-ai/deepseek-harness · 4,261 tokens