No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by Playa-Cyrene · MCP Server · ★ 606
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is Cyrene-Agent safe to install? View the security audit →
Cyrene-Agent English | 中文 Cyrene-Agent 是一个以《崩坏：星穹铁道》昔涟为核心角色的 Windows Live2D AI 桌面伴侣。 基于 Electron + TypeScript 开发的桌面端 Live2D 智能对话 Agent。 项目围绕昔涟（Cyrene）的角色设定，结合自研CyreneHarness+DMAE 记忆引擎， 将角色化聊天、个性化记忆、语音交互、工具调用与多平台接入整合在同一个桌面 Agent 中， 支持日常聊天（Chat）、辅助工作（Work）、代码协作（Code）、学习陪伴（Learn）四种对话模式。 ✨ 速览 🌸 趣味桌面陪伴 — Live2D 角色常驻桌面，支持表情、动作、状态、心情、气泡互动与智能表情包 💬 日常聊天（Chat） — 专注角色化交流，结合会话历史、用户风格与长期记忆自然回应 🛠️ 辅助工作（Work） — 通用任务会话，支持联网搜索、文件处理、文档生成、生活服务等工具的串联调用，由 CyreneHarness 主循环统一调度 💻 代码协作（Code） — 绑定可信代码目录，提供 LSP 语义查询（定义/引用/悬停/符号/诊断）与受限的读写改命令执行，安全边界由 Harness 的权限审批（Permission Policy）与 Execution Policy 统一把关 📚 学习陪伴（Learn） — 绑定 Obsidian Vault，陪伴用户理解材料、整理笔记、生成练习与维护进 🧠 个性化记忆 — L0 / L1 / L2 分层记忆，结合自研记忆头像+Worldbook+沉淀长期互动 🌱 条目生命周期 — 自研DMAE算法（v4.0未实现最新v5.1）负责管理prompt在上下文中的生命周期 🔊 语音交互 — 集成 TTS、ASR 与语音通话，让昔涟能够听见并回应用户 🧰 丰富工具生态 — 覆盖联网搜索、文件处理、文档生成、生活服务、音乐与 MCP 扩展 🧩 插件系统 — 本地插件包扩展 AI 工具、聊天渠道、自有窗口与语音输入，配套 npm SDK 与开发指南 🔌 多模型厂商适配 — 针对不同厂商提供分级 Structured Output 与 Function Calling 兼容方案 🎨 个性化外观 — 支持多套界面风格、主题外观与聊天字体选择 📱 多平台接入 — 支持桌面端、飞书、微信 iLink 与 QQ（NapCat / OneBot 11），共享角色能力与对话体验 🌙 主动聊天 — 根据时间、状态与用户偏好主动发起交流，并支持多渠道定向投递 ⚙️ CyreneHarness 核心引擎 等需要工具调用的会话模式，全部跑在 CyreneHarnes
| Stars | 606 |
| Forks | 85 |
| Language | TypeScript |
| Category | MCP Server |
| License | MIT |
| Quality Score | 63.670024640297/100 |
| Open Issues | 15 |
| Last Updated | 2026-09-19 |
| Created | 2026-07-06 |
| Platforms | mcp, node |
| Est. Tokens | ~21k |
These tools work well together with Cyrene-Agent for enhanced workflows:
Looking for a Cyrene-Agent alternative? If you're comparing Cyrene-Agent with other mcp server tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
An open-source desktop AI agent built around Cyrene’s persona and powered by the self-developed Cyrene_Harness
🍙 A personal AI agent & local memory hub for all AI agents, gives every AI one shared, fully controlled memo
Local-first personal agentic OS and everything app for coding, knowledge work, web design, automations, and ar
Agent harness and tookit for building AI agents and agentic applications. CLI and SDKs included
Aser is a lightweight, self-assembling AI Agent frame.
Persistent memory for AI coding agents
Explore other popular mcp server tools:
Cyrene-Agent is An open-source desktop AI agent built around Cyrene’s persona and powered by the self-developed Cyrene_Harness framework. It combines immersive character chat with practical Agent capabilities for dai. It is categorized as a MCP Server with 606 GitHub stars.
Cyrene-Agent is primarily written in TypeScript. It covers topics such as agent, ai-agent, ai-assistants.
You can find installation instructions and usage details in the Cyrene-Agent GitHub repository at github.com/Playa-Cyrene/Cyrene-Agent. The project has 606 stars and 85 forks, indicating an active community.
Cyrene-Agent is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to Cyrene-Agent on Agent Skills Hub include Cyrene-Agent, memmy-agent, CoWork-OS. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: