Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/xkzs2007-galaxyos
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/galaxyos
What's inside
- Skill name
galaxyos- Description
- GalaxyOS 认知增强型 AI Agent 桌面引擎 v0.3.0 Cognitive Nexus
- License
- MIT
- Instructions
- 268 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Agent Skills spec issues
- name: "galaxyos" does not match folder "GalaxyOS-desktop"
Files (200)
- .cnb.yml
- .dockerignore
- .gitattributes
- CODEOWNERS · .github/
- ci.yml · .github/workflows/
- lint.yml · .github/workflows/
- release.yml · .github/workflows/
- security.yml · .github/workflows/
- test.yml · .github/workflows/
- .gitignore
- .gitleaks.toml
- verified_memories.jsonl · .learnings/
- .pre-commit-config.yaml
- AGENTS.md
- CHANGELOG.md
- CONTEXT.md
- CONTRIBUTING.md
- Dockerfile.buildkit
- Dockerfile.ci
- Dockerfile.kernel
- GalaxyOS_模块技术说明文档_v0.3.0.docx
- LICENSE
- Makefile
- README.md
- SKILL.md
- UBIQUITOUS_LANGUAGE.md
- VERSION
- VERSIONING.md
- backup_metadata.json · backups/backup_20260411_142857_manual/
- monitor_config.json · backups/backup_20260411_142857_manual/config/
- performance_config.json · backups/backup_20260411_142857_manual/config/
- priority_config.json · backups/backup_20260411_142857_manual/config/
- system_config.json · backups/backup_20260411_142857_manual/config/
- fusion_registry.json · backups/backup_20260411_142857_manual/inventory/
- skill_registry.json · backups/backup_20260411_142857_manual/inventory/
- backup_manager.py · backups/
- ci-check-version.py · bin/
- dsl_bridge_mappings.json · config/
- evolution_triggers.json · config/
- integration_config.json · config/
- llm_config.example.json · config/
- module_dependencies.json · config/
- monitor_config.json · config/
- performance_config.json · config/
- priority_config.json · config/
- system_config.json · config/
- prompt_integration.py · core/
- unified_vector_store.py · core/
- rccam_classifier_v1.pkl · data/
- CMakeLists.txt · desktop-native/
- CPackPlatformConfig.cmake · desktop-native/
- i18n_bridge.h · desktop-native/include/
- native_config.h · desktop-native/include/
- native_event_bus.h · desktop-native/include/
- native_ipc_channel.h · desktop-native/include/
- native_logger.h · desktop-native/include/
- native_process_manager.h · desktop-native/include/
- native_sse_client.h · desktop-native/include/
- galaxyos.desktop · desktop-native/scripts/
- post-install.sh · desktop-native/scripts/
- …and 140 more
SKILL.md (first lines)
# GalaxyOS v0.3.0 Cognitive Nexus
> **定位**:GalaxyOS 认知增强型 AI Agent 桌面引擎 v0.3.0 Cognitive Nexus
> **架构**:C++ 桌面壳 (EUI-NEO GPU 直渲) + GalaxyOS Python 内核 (MCP Server + AgentCore Bridge)
> **最新特性**:v0.3.0 双进程架构落地 + EUI-NEO 原生渲染 + 三级渲染降级链
---
## 🎯 核心能力
GalaxyOS 作为独立桌面引擎,采用双进程架构运行:
| 进程 | 技术栈 | 职责 |
|------|--------|------|
| **C++ 桌面壳** | EUI-NEO C++ SDK + GLFW + OpenGL | GPU 直渲、窗口管理、IPC、SSE、托盘 |
| **Python 内核** | MCP Server + openJiuwen agent-core | 认知引擎、记忆系统、技能编排、工具服务 |
### 7 大能力模块
1. **液态神经记忆** — LTC 突触 / CfC 推理 / NCP 神经电路策略 / SSM 状态预测 / 仿生遗忘曲线
2. **DAG 上下文管理** — SQLite 持久化 / 摘要节点回溯 / 时间衰减排序 / 多粒度优先级
3. **COSPLAY 技能自演化** — 从执行轨迹学习技能合约 → ProtoSkill → 毕业为成熟 Skill
4. **EUI-NEO 原生渲染** — C++ GPU 直渲 / TokUI DSL → EUI-NEO 组件映射 / 流式增量构建
5. **R-CCAM 认知循环** — Retrieval → Cognition → Control → Action → Memory 五阶段结构化
6. **MultiAgent 协同** — 5 角色(searcher/analyst/architect/critic/summarizer)+ 公告板 + 蒸馏 + 交叉验证
7. **三级渲染降级** — `eui_native` (EUI-NEO GPU 直渲) → `webview_dom` → `plain_text`(不可逆)
---
## 🏗️ 架构
```
┌─────────────────────────────────────────────────────────────┐
│ C++ Desktop Shell (EUI-NEO GPU 直渲) │
│ GLFW 窗口 / OpenGL 渲染 / NativeIPCChannel / NativeSSE │
│ NativeProcessManager / NativeEventBus / I18nBridge │
└─────────────────────────────────────────────────────────────┘
│
HTTP IPC (cpp-httplib)
SSE (:8765)
│
┌─────────────────────────────────────────────────────────────┐
│ GalaxyOS Python 内核 — MCP Server │
├─────────────────────────────────────────────────────────────┤
│ MCP Server 入口 (mcp_server_entry) │
│ AgentCore Bridge (openJiuwen agent-core @v0.1.16) │
│ DSL Bridge (TokUI DSL → EUI-NEO 渲染指令) │
├─────────────────────────────────────────────────────────────┤
│ 1. 液态神经核心 (LTC/CfC/NCP/SSM) │
│ 2. DAG 上下文 (SQLite + 摘要回溯 + 时间衰减) │
│ 3. COSPLAY 适配 (边界检测 + 合约学习 + 毕业) │
│ 4. R-CCAM 认知循环 (5 阶段 + 元认知调节) │
│ 5. MultiAgent 编排 (5 角色 + 公告板 + 蒸馏) │
│ 6. 防幻觉 10 重检测 (Self-RAG/CRAG/CoVe) │
│ 7. ONNX Embedding (bge-small-zh-v1.5-ONNX) │
└─────────────────────────────────────────────────────────────┘
```
---
## 🔌 必需的配置
配置文件位于 `$GALAXYOS_HOME/config/` 目录,核心配置项:
```json
{
"mode": "desktop",
"mcp_server": {
"host": "127.0.0.1",
"port": 8765
},
"rendering": {
"fallback_chain": ["eui_native", "webview_dom", "plain_text"]
},
"agent_core": {
"backend": "openjiuwen",
"version": "0.1.16"
}
}
```
环境变量 `GALAXYOS_HOME` 指向 GalaxyOS 数据目录,未设置时默认 `~/.galaxyos/`。
---
## 📊 生命周期
| 阶段 | 触发时机 | 用途 |
|------|----------|------|
| `mcp_server_start` | MCP Server 启动 | 初始化引擎、注册工具、启动 SSE |
| `mcp_server_stop` | MCP Server 停止 | 统一关闭所有组件、持久化状态 |
| `agent_core_init` | agent-core 初始化 | 加载记忆系统、技能库、认知循环 |
| `agent_core_shutdown` | agent-core 关闭 | 保存记忆、flush engram、关闭连接 |
| `before_tool_call` | 工具调用前 | 记录调用前状态给 BoundaryDetector |
| `after_tool_call` | 工具调用后 | 幂等捕获结果,更新 Skill Bank + engram + DAG |
| `before_compaction` | 压缩前 | 高价值上下文持久化到 engram + DAG |
| `after_compaction` | 压缩后 | 向量索引同步 |
| `render_fallback` | 渲染降级 | eui_native → webview_dom → plain_text 逐级降级 |
---
## 🛠️ 20 个 MCP 工具
| 工具 | 用途 |
|------|------|
| `galaxy_pool` | GalaxyPool 状态查询 |
| `galaxy_rccam_progress` | R-CCAM 实时进度 |
| `galaxy_recall` | 深度语义记忆检索 |
| `galaxy_health` | 系统健康检查 |
| `galaxy_vector_info` | 向量计算能力 |
| `galaxy_events` | 事件日志查询 |
| `galaxy_store` | 记忆存储 |
| `galaxy_verify` | 幻觉验证 |
| `galaxy_rccam` | R-CCAM 认知循环 |
| `galaxy_save_memory` | 记忆持久化 |
| `galaxy_compile_skill` | Skill 编译(SkVM) |
| `galaxy_asset_search` | KnowledgeAsset 搜索 |
| `galaxy_asset_register` | KnowledgeAsset 注册 |
| `galaxy_dag_status` | DAG 上下文状态 |From xkzs2007/GalaxyOS-desktop, MIT.
More from xkzs2007/GalaxyOS-desktop
- 0.569score
darwin-skill
xkzs2007
Darwin Skill (达尔文.skill): autonomous skill optimizer inspired by Karpathy's autoresearch.
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
criticism-self-criticism
xkzs2007
触发:当一项工作已经完成、进入阶段验收、收到批评反馈,或反复出现同类错误需要系统纠偏时调用;常见信号包括 review、audit、retrospective、quality check、纠错与复盘。 English: Trigger after delivery or at a review checkpoint…
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
arming-thought
xkzs2007
触发:在每次新的顶层对话开始时自动调用,用于建立“实事求是”的总原则,并在明确适用时为后续任务选择下游 skill;如果你是被派遣执行单一具体任务的子 agent,则跳过此 skill。 English: Trigger at the start of each new top-level conversation…
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
concentrate-forces
xkzs2007
触发:当多个任务同时争夺时间、注意力、算力或预算,必须确定主攻方向并停止分散用力时调用;常见信号包括优先级过多、资源紧张、推进分散、需要决定先做什么。 English: Trigger when limited resources are being split across too many tasks and…
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
contradiction-analysis
xkzs2007
触发:当问题复杂、存在多个冲突因素、优先级不清,或你不知道应该先解决什么时调用;常见信号包括 trade-off、瓶颈、根因不明、主次不清、多个问题互相牵制。 English: Trigger when a problem contains competing forces, unclear priorities…
SkillBFreshness: B★ 3Claude Code +9 - 0.561score
autonomous-tasks
xkzs2007
Self-driven AI worker.
SkillBFreshness: B★ 3Claude Code +9 - 0.560score
brain
xkzs2007
Personal knowledge base for capturing and retrieving information about people, places, restaurants, games, tech, events, media, ideas, and organizations.
SkillBFreshness: B★ 3Claude Code +9 - 0.558score
best-minds
xkzs2007
模拟器思维:不问"你怎么看",而是问"世界上谁最懂这个?TA 会怎么说?"。触发词:最强大脑、顶级专家、世界级、best minds、谁最懂这个
SkillBFreshness: B★ 3Claude Code +9 - 0.558score
caveman
xkzs2007
超压缩通信模式。砍掉约 75% 的废话 token,保留全部技术信息。触发词:caveman 模式、简洁回复、省 token。
SkillBFreshness: B★ 3Claude Code +9 - 0.555score
critical-thinking
xkzs2007
批判性思维。评估证据、识别逻辑谬误、检验假设。触发词:评估、论证、证据、逻辑谬误、批判性思维。
SkillBFreshness: B★ 3Claude Code +9 - 0.555score
analogical-thinking
xkzs2007
类比思维。从其他领域的成功模式中借鉴解决方案。触发词:类比、借鉴、像...一样、跨领域。
SkillBFreshness: B★ 3Claude Code +9 - 0.555score
backward-thinking
xkzs2007
逆向思维。从目标倒推起点,识别关键路径。触发词:倒推、逆向、从终点出发、逆向思维。
SkillBFreshness: B★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 0
- Open issues
- 0
- Last commit
- Aug 1, 2026 (2 months ago)
- Last release
- Aug 1, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/xkzs2007/GalaxyOS-desktop
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from xkzs2007
- 0.569score
darwin-skill
xkzs2007
Darwin Skill (达尔文.skill): autonomous skill optimizer inspired by Karpathy's autoresearch.
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
criticism-self-criticism
xkzs2007
触发:当一项工作已经完成、进入阶段验收、收到批评反馈,或反复出现同类错误需要系统纠偏时调用;常见信号包括 review、audit、retrospective、quality check、纠错与复盘。 English: Trigger after delivery or at a review checkpoint…
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
arming-thought
xkzs2007
触发:在每次新的顶层对话开始时自动调用,用于建立“实事求是”的总原则,并在明确适用时为后续任务选择下游 skill;如果你是被派遣执行单一具体任务的子 agent,则跳过此 skill。 English: Trigger at the start of each new top-level conversation…
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
concentrate-forces
xkzs2007
触发:当多个任务同时争夺时间、注意力、算力或预算,必须确定主攻方向并停止分散用力时调用;常见信号包括优先级过多、资源紧张、推进分散、需要决定先做什么。 English: Trigger when limited resources are being split across too many tasks and…
SkillBFreshness: B★ 3Claude Code +9 - 0.564score
contradiction-analysis
xkzs2007
触发:当问题复杂、存在多个冲突因素、优先级不清,或你不知道应该先解决什么时调用;常见信号包括 trade-off、瓶颈、根因不明、主次不清、多个问题互相牵制。 English: Trigger when a problem contains competing forces, unclear priorities…
SkillBFreshness: B★ 3Claude Code +9 - 0.561score
autonomous-tasks
xkzs2007
Self-driven AI worker.
SkillBFreshness: B★ 3Claude Code +9
Cite this page
galaxyos skill. Agent Mart. https://agentmart.dev/skills/xkzs2007-galaxyos (last updated 2026-10-24)