Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/chyinan/kokoro-engine/claude-mdgit clone --depth 1 https://github.com/chyinan/Kokoro-EngineWrote 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.
[](https://agentmods.dev/instructions/chyinan/kokoro-engine/claude-md)<a href="https://agentmods.dev/instructions/chyinan/kokoro-engine/claude-md"><img src="https://agentmods.dev/badge/instructions/chyinan/kokoro-engine/claude-md.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.04903 | $0.04903 |
| Opus 5 | $0.02452 | $0.02452 |
| Sonnet 5 | $0.00981 | $0.00981 |
| Haiku 4.5 | $0.00490 | $0.00490 |
Grade A, and why
Kokoro-Engine 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 503 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Git 规范
- Commit 信息必须使用英文，便于国际开发者协作
- Commit 格式遵循 Conventional Commits：
type: description（如feat: add voice interrupt support、fix: resolve TTS cache invalidation） - Commit之后请自动推送到GitHub
项目概述
Kokoro Engine 是一个基于 Tauri v2 的跨平台桌面虚拟角色沉浸交互引擎。前端使用 React + TypeScript，后端使用 Rust。整合了 Live2D、LLM、TTS、STT、Vision、ImageGen、MCP 等多种 AI 能力，并通过 MOD 系统支持 UI 和功能扩展。
常用命令
# 安装依赖
npm install
# 启动开发环境（前端 + Tauri 后端）
npm run tauri dev
# 构建发行版
npm run tauri build
# 前端测试（Vitest）
npm run test
# 运行单个前端测试文件
npx vitest run src/ui/mods/ModMessageBus.test.ts
# 后端测试（Rust）
cd src-tauri && cargo test
# 运行单个后端测试
cd src-tauri && cargo test test_name
# 前端类型检查
npx tsc --noEmit
# Rust 编译检查
cd src-tauri && cargo check
# Rust lint
cd src-tauri && cargo clippy
架构
双层架构，通过 Tauri IPC 通信：
Frontend (React + TS) ←→ kokoro-bridge.ts (Typed IPC) ←→ Backend (Rust/Tauri)
前端 (src/)
core/— 服务初始化、单例服务实例、共享类型定义features/live2d/— Live2D 渲染（PixiJS 6 + Cubism SDK），包含表情驱动、口型同步、动作控制ui/layout/— 声明式布局引擎，JSON 配置驱动的 grid/layer/component 树ui/registry/— 组件注册表（单例模式），按名称注册和解析组件，支持 MOD 注入ui/theme/— 主题系统，CSS 变量注入，支持 MOD 主题覆盖ui/mods/— MOD UI 组件，iframe 沙箱通信lib/kokoro-bridge.ts— 所有 Tauri IPC 调用的类型化封装层，是前后端通信的唯一入口
后端 (src-tauri/src/)
commands/— 20 个 IPC 命令处理模块（chat、tts、stt、llm、vision、imagegen、mcp、mods、telegram 等）ai/— AI 编排器：上下文管理（context.rs）、情感系统（emotion.rs，持久化）、记忆系统（memory.rs）、自主行为（好奇心、主动发言、闲置动作）llm/— LLM 适配器（OpenAI 兼容接口、Ollama），支持多 Provider + 预设切换、Fast/Smart/Cheap 模型路由tts/— TTS 服务，6 个提供商（GPT-SoVITS、VITS、OpenAI、Azure、ElevenLabs、Browser TTS），trait 化插件架构stt/— 语音识别（Whisper、OpenAI）vision/— 屏幕截图 + VLM 分析（xcap 截图，像素差异检测）imagegen/— 图片生成（Stable Diffusion WebUI、OpenAI DALL-E、Google Gemini）mcp/— Model Context Protocol 客户端（stdio + Streamable HTTP 传输，JSON-RPC 2.0）mods/— MOD 系统：manifest 解析、mod://协议、QuickJS (ES2020) 脚本沙箱telegram/— Telegram Bot 远程交互（teloxide 长轮询，Chat ID 白名单，文字/语音/图片消息桥接）actions/— 动作注册表，用于 LLM tool calling
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.
- 5d ago First seen · 503 lines · 4,903 tokens per session scan A a673a49f0bf2
Kokoro-Engine CLAUDE.md is an instructions file published in the GitHub repository chyinan/Kokoro-Engine (144 stars, last pushed 2d ago), licensed MIT. It adds 4,903 tokens to every session, about $0.0245 per session on Opus 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 2026-08-30.
Other instructions, from other repositories
own-planner frontend.instructions.md
Instructions for am-space/own-planner, covering frontend instructions, technology, coding rules and change preferences.
frontend-stack-blueprint CLAUDE.md
Instructions for barrosohub/frontend-stack-blueprint, covering frontend stack blueprint v2.0.0, the stack (use for every project), architecture (mandatory — apply to all code), universal governance protocol (technology-agnostic) and production reliability (mandatory by applicable profile).
Little-Claude CLAUDE.md
Instructions for qingyu321/Little-Claude, covering claude.md, project overview, development commands, install dependencies and frontend only (vite dev server).
thunderbolt AGENTS.md
AGENTS.md instructions for thunderbird/thunderbolt, covering core principles, typescript & code style, tooling & libraries, react patterns and useeffect discipline.
copilot-instructions
This file describes the coding style and guidelines for the project, which will be used by GitHub Copilot to generate code suggestions that align with the project's standards.
react-seed AGENTS.md
Instructions for guokaigdg/react-seed, covering agents.md, 目录, 1. 项目概览, 2. 技术栈 and 3. 目录结构.