[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-83302":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":14,"stars7d":16,"stars30d":16,"stars90d":15,"forks30d":15,"starsTrendScore":17,"compositeScore":18,"rankGlobal":10,"rankLanguage":10,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":20,"hasPages":20,"topics":22,"createdAt":10,"pushedAt":10,"updatedAt":23,"readmeContent":24,"aiSummary":10,"trendingCount":15,"starSnapshotCount":15,"syncStatus":25,"lastSyncTime":26,"discoverSource":27},83302,"gliding_horse","doiito\u002Fgliding_horse","doiito","Gliding Horse is a multi-agent orchestration framework built in Rust that supports PDCA scheduling and knowledge graph-based agents, with comprehensive Chinese documentation, and is suitable for building enterprise-level AI agent systems.","",null,"Rust",71,10,1,0,17,12,55.32,"MIT License",false,"main",[],"2026-06-12 04:01:40","# Gliding Horse Agent OS (流马智能体操作系统)\n\u003Cdiv align=\"center\">\n\n![Gliding Horse Logo](assets\u002Flogo.jpg)\n\n**An Industrial-Grade AI Agent Operating System Built in Rust**  [![Star on GitHub](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdoiito\u002Fgliding_horse?style=flat)](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse)\n\n*Inspired by Zhuge Liang's Wooden Ox and Gliding Horse — Ancient Ingenuity Meets Modern AI*\n\n[![Rust](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRust-2021-orange.svg)](https:\u002F\u002Fwww.rust-lang.org\u002F)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-MIT-blue.svg)](LICENSE)\n[![gRPC](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FgRPC-Protocol-green.svg)](https:\u002F\u002Fgrpc.io\u002F)\n[![Knowledge Graph](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FKnowledge%20Graph-Oxigraph-purple.svg)](https:\u002F\u002Foxigraph.org\u002F)\n[![Release](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Frelease\u002Fdoiito\u002Fgliding_horse)](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse\u002Freleases)\n\n---\n\n[**English**](README.md) · [**中文**](README.zh.md) · [**Design Detail →**](docs\u002FDESIGN_DETAIL.md)\n\n\u003C\u002Fdiv>\n\n---\n\n## What Is Gliding Horse?\n\nAn **AI agent operating system** built in Rust that orchestrates multiple agents via the PDCA cycle, enabling coordinated, auditable, and self-improving systems. — much like how Zhuge Liang's **Wooden Ox and Gliding Horse** revolutionized logistics by harnessing mechanical power across treacherous terrain.\n\n> \"We don't just build agents; we build the **infrastructure that harnesses their collective intelligence**.\"\n\n### Core Architecture\n\n| Layer | Technology | Role |\n|-------|-----------|------|\n| **Core Coordination** (Rust) | `PDCA cycle` · `5W2H ontology` · `EventBus` | Agent orchestration & lifecycle |\n| **Memory System** | `L0: Sled+Qdrant` · `L2: Oxigraph` · `MESI coherence` | 5-layer hierarchical memory |\n| **Data Bus** | `JSON-LD 1.1` · `@id\u002F@type\u002F@context` · `Named Graphs` | Universal interoperability |\n| **Knowledge Graph** | `Oxigraph RDF` · `SPARQL 1.1` · `Code AST` | Cross-subsystem unified store |\n| **Skill Graph** | `RDF` · `7.5k LOC` · `Self-evolving` | Dynamic cognitive network |\n| **Perception Engine** | `10 triggers` · `Anomaly dedup` · `5W2H constraint check` | Proactive monitoring |\n| **Gateway** | `gRPC` · `HTTP (OpenAI-compatible)` · `MCP` | Production interface |\n\n---\n\n## 📖 The Story: From Ancient Wisdom to Modern Intelligence\n\nIn the turbulent era of the Three Kingdoms (220–280 AD), the legendary strategist **Zhuge Liang** (诸葛亮), chancellor of the Shu Han state, faced a critical challenge: how to transport supplies efficiently through the treacherous mountain paths of Sichuan during his Northern Expeditions. Traditional wheeled carts struggled on narrow trails; human porters exhausted quickly.\n\nHis solution — the **Wooden Ox (木牛)** and **Gliding Horse (流马)** — were autonomous transport devices that could navigate difficult terrain with minimal human guidance. These mechanical wonders were not merely tools; they represented a paradigm shift — **autonomous systems that extended human capability**.\n\n### Bridging Past and Present\n\nJust as the Gliding Horse served as an **intelligent harness** for transporting supplies across impossible terrain, **Gliding Horse Agent OS** serves as an **intelligent harness for AI agents**:\n\n| Ancient Innovation | Modern Implementation |\n|-------------------|----------------------|\n| **Autonomous Transport** | Self-directing agent workflows |\n| **Terrain Adaptation** | Dynamic complexity handling (7 levels) |\n| **Load Distribution** | Parallel agent execution |\n| **Minimal Guidance** | Proactive anomaly detection |\n| **Mechanical Reliability** | Rust's memory safety guarantees |\n\n> *\"The wise adapt their methods to circumstances, just as water shapes its course according to the ground over which it flows.\"*  \n> — **Zhuge Liang**\n\nThis ancient wisdom guides our design: **flexible orchestration that adapts to task complexity**, rather than rigid frameworks that force tasks into predefined molds.\n\n---\n\n## 🖥️ Software Engineering Team — The Flagship Application\n\nThe **Software Engineering Team** app demonstrates the full power of Gliding Horse — a federated architecture where multiple AI agents collaborate on real-world software engineering tasks.\n\n![Dashboard](assets\u002Fdashboard.JPG)\n*Center dashboard — project oversight, agent status, pipeline progress*\n\n\u003Cdiv align=\"center\">\n  \u003Ctable>\n    \u003Ctr>\n      \u003Ctd>\u003Cimg src=\"assets\u002Fproject.JPG\" alt=\"Project Management\" width=\"400\"\u002F>\u003Cbr\u002F>\u003Cem>Project lifecycle management\u003Cbr\u002F>from req → design → code → review → deploy\u003C\u002Fem>\u003C\u002Ftd>\n      \u003Ctd>\u003Cimg src=\"assets\u002Fpipeline.JPG\" alt=\"Pipeline Visualization\" width=\"400\"\u002F>\u003Cbr\u002F>\u003Cem>Multi-stage SDLC pipeline\u003Cbr\u002F>with real-time status tracking\u003C\u002Fem>\u003C\u002Ftd>\n    \u003C\u002Ftr>\n  \u003C\u002Ftable>\n\u003C\u002Fdiv>\n\n![VS Code Plugin](assets\u002Fvscode_plugin.JPG)\n*VS Code Plugin — chat panel, graph view, and task panel for real-time agent collaboration*\n\n### Architecture: Center + Edge Federation\n\n```mermaid\nflowchart TB\n    subgraph VS[\"VS Code Plugin (TypeScript)\"]\n        direction LR\n        CHAT[\"Chat Panel\"]\n        GRAPH_V[\"Graph View\"]\n        TASK_P[\"Task Panel\"]\n    end\n\n    subgraph EDGE[\"Edge Daemon (Rust · axum)\"]\n        API_EDGE[\"API Server\u003Cbr\u002F>ws \u002F chat \u002F health\"]\n        AGENT_CORE[\"Agent Core\u003Cbr\u002F>SupervisorAgent · DoAgent · LLM Client\"]\n        DOCKER[\"Docker Sandbox\u003Cbr\u002F>Safe execution · Compile \u002F Test\"]\n        SYNC_EDGE[\"Sync Layer\u003Cbr\u002F>Heartbeat · gRPC · JWT Auth\"]\n        GRAPH_EDGE[\"Graph Layer\u003Cbr\u002F>Local Store (sled) · Delta Sync\"]\n        \n        API_EDGE --- AGENT_CORE\n        AGENT_CORE --- DOCKER\n        API_EDGE --- SYNC_EDGE\n        AGENT_CORE --- GRAPH_EDGE\n    end\n\n    subgraph CTR[\"Center (Go · Gin)\"]\n        API_CTR[\"HTTP API\u003Cbr\u002F>\u002Fapi\u002Fv1\u002F* · \u002Fws\"]\n        TEMPORAL[\"Temporal Workflow\u003Cbr\u002F>Orchestrator\"]\n        AGENT_MGR[\"Agent Manager\u003Cbr\u002F>Register · Heartbeat · Dispatch\"]\n        EXEC[\"Executors\u003Cbr\u002F>req → design → coding → review → test → cicd → deploy\"]\n        STORE_CTR[\"Store\u003Cbr\u002F>SQLite · gRPC Client · Graph Sync\"]\n        \n        API_CTR --- TEMPORAL\n        API_CTR --- AGENT_MGR\n        TEMPORAL --- EXEC\n        AGENT_MGR --- STORE_CTR\n    end\n\n    VS \u003C-->|\"WebSocket \u002F REST\"| EDGE\n    EDGE \u003C-->|\"gRPC + REST\"| CTR\n```\n\n**Key Design Patterns:**\n- **Center (Go)**: Workflow orchestration via Temporal, project CRUD, agent registry, graph sync\n- **Edge (Rust)**: Local LLM execution, Docker sandbox, VS Code WebSocket bridge\n- **VS Code Plugin**: Developer UI with real-time agent awareness\n\n---\n\n## 🖥️ Gliding Code — The Terminal AI Assistant\n\n**Gliding Code** is a terminal-based AI coding assistant that brings the power of Gliding Horse's knowledge graph and agent orchestration directly into your command line — no IDE required.\n\n![Gliding Code Demo](assets\u002Fscreenshot.gif)\n\n![Knowledge Graph in Action](assets\u002Fgliding_code_kg.JPG)\n*Knowledge graph visualization — real-time entity relationships, code structure understanding, and cross-subsystem awareness powered by Oxigraph RDF*\n\n![Completed Programming Task](assets\u002Fgliding_code.JPG)\n*Task completion interface — AI agent successfully analyzing and solving a programming task with full traceability*\n\n---\n\n## 🚀 Quick Start\n\nChoose your path — **download and run** the pre-built terminal AI assistant (zero dependencies), or **build from source** for the full Software Engineering Team.\n\n### Option A: Download & Run — Gliding Code\n\nNo dependencies required. Just download, extract, and run:\n\n| Platform | Download |\n|----------|----------|\n| Linux (x86_64, musl) | [`glidingcode-x86_64-unknown-linux-musl.tar.gz`](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse\u002Freleases) (13.9 MB) |\n| Linux (aarch64, musl) | [`glidingcode-aarch64-unknown-linux-musl.tar.gz`](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse\u002Freleases) (12.9 MB) |\n| macOS (Apple Silicon) | [`glidingcode-aarch64-apple-darwin.tar.gz`](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse\u002Freleases) (12.1 MB) |\n| Windows (x86_64) | [`glidingcode-x86_64-pc-windows-msvc.zip`](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse\u002Freleases) (11.6 MB) |\n\n```bash\n# Linux \u002F macOS\ntar xzf glidingcode-*.tar.gz\n.\u002Fglidingcode --help\n\n# Windows (PowerShell)\nExpand-Archive glidingcode-x86_64-pc-windows-msvc.zip .\n.\\glidingcode.exe --help\n```\n\n> All Linux builds are **fully statically linked** (musl) — no runtime dependencies required.\n\nSet your API key and start using it:\n\n```bash\nexport DEEPSEEK_API_KEY=\"sk-...\"        # Linux \u002F macOS\n# or\nset DEEPSEEK_API_KEY=\"sk-...\"            # Windows (cmd)\n# or\n$env:DEEPSEEK_API_KEY=\"sk-...\"           # Windows (PowerShell)\n\n# Alternatively, use any OpenAI-compatible provider:\nexport AGENT_OS_GATEWAY_API_KEY=\"sk-...\"\nexport AGENT_OS_GATEWAY_API_URL=\"https:\u002F\u002Fyour-endpoint\u002Fv1\"\n\n# Run an interactive session (Linux\u002FmacOS: .\u002Fglidingcode, Windows: .\\glidingcode)\n.\u002Fglidingcode\n\n# Or run a one-shot task\n.\u002Fglidingcode \"Explain how Rust's borrow checker works\"\n```\n\n### Option B: Full Setup — Software Engineering Team\n\nBuild the complete multi-agent system from source (requires Rust + Go + Docker).\n\n#### Prerequisites\n\n- **Rust** 1.75+ · **Go** 1.25+ · **Docker** · **Temporal Server**\n- LLM API key (OpenAI-compatible)\n\n#### 1. Clone & Configure\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse.git\ncd gliding_horse\u002Fapps\u002Fsoftware_engineering_team\n\ncp center\u002Fconfig.yaml center\u002Fconfig.local.yaml\n# Edit your LLM keys, Temporal host, etc.\n```\n\n#### 2. Start the Center\n\n```bash\ncd center\ngo run .\u002Fcmd\u002Fserver\u002F...     # API server on :8080\ngo run .\u002Fcmd\u002Fworker\u002F...     # Temporal worker\n```\n\n#### 3. Start the Edge Daemon\n\n```bash\ncd edge\u002Fdaemon\ncargo run -- daemon start   # Agent daemon on :7890\n```\n\n#### 4. Open VS Code\n\nInstall the plugin from `edge\u002Fvscode\u002F` and connect to the daemon — you now have an AI software engineering team at your fingertips.\n\n#### Or Use the API Directly\n\n```bash\ncurl http:\u002F\u002Flocalhost:8080\u002Fapi\u002Fv1\u002Fprojects \\\n  -X POST -H \"Content-Type: application\u002Fjson\" \\\n  -d '{\"name\":\"My Project\",\"description\":\"Build a microservice\"}'\n```\n\n---\n\n## 🔧 Key Highlights\n\n1. **Generalized PDCA — 7-Level Adaptive Execution**  \n   Dynamically selects from 7 complexity levels (L0 instant → L5 recursive → L6 emergency) via 5W2H metadata. One engine handles everything from instant queries to multi-week projects — no rigid workflows.\n\n2. **CPU Cache-Inspired Memory — 5 Layers + MESI Coherence**  \n   First-ever application of CPU cache coherence to multi-agent memory. L0 disk → L1 context → L2 Oxigraph RDF → L3 SPARQL projection. Intelligent prefetching reduces perceived latency by 90%. Solves context explosion and shared memory inconsistency.\n\n3. **JSON-LD Universal Data Bus — W3C-Standard Interoperability**  \n   `@context` duck-typing eliminates field name conflicts between skills. `@id` enables zero-cost cross-agent entity merging. `@graph` named graphs allow conflict-free parallel writes. Turns interoperability hell into plug-and-play.\n\n4. **Self-Evolving Skill Graph — Cognitive Network**  \n   7,500+ LOC dynamic network with 6 semantic link types (Prerequisite, Composition, Related, etc.). AA creates knowledge fragments and new links after each task. `\u002Flearn` and `\u002Freduce` mechanisms enable autonomous skill acquisition.\n\n5. **Universal Knowledge Graph — Unified Cognitive Backbone**  \n   All subsystems (skills, memories, tasks, code knowledge) share a single Oxigraph RDF store via named graphs, enabling cross-subsystem SPARQL joins. Code ASTs parsed by tree-sitter are automatically converted to RDF triples and linked into the same graph. A single `@id` ensures consistent entity identity across all contexts — no silos, no duplication.\n\n6. **5W2H Dimension-Level Audit — Precision Rollback**  \n   CA audits each of the 7 dimensions independently. What\u002FWhy fail → re-analyze. How\u002FWhere fail → re-plan. When\u002FHowMuch fail → conditional pass. No more black-box \"PASS\u002FFAIL\" — you know exactly what went wrong.\n\n7. **Proactive Perception Engine — Catch Failures Before They Happen**  \n   10 execution triggers with 60-second anomaly deduplication. Monitors deadline violations, budget overruns (>80% tokens), role mismatches, environment conflicts. Auto-escalates to human when needed.\n\n8. **Micro-Tool System — Tame Large Outputs**  \n   Results >8KB auto-generate conversational micro-tools (e.g., \"search_in_results\"). Transforms unwieldy 50KB+ outputs into interactive, queryable artifacts within the LLM context.\n\n9. **MCP Integration — One Protocol to Connect Them All**  \n   Standard Model Context Protocol connects GitHub, Slack, Jira, and any MCP-compatible server. Dynamic tool discovery at runtime. No more custom integrations for every external service.\n\n10. **Checkpoint & Recovery — Crash-Proof Long-Running Tasks**  \n    Session state snapshots at critical points. Full restoration on crash without context loss. Enables hour\u002Fday-long agent tasks and post-mortem replay debugging.\n\n11. **Center + Edge Federation — Local Autonomy, Global Orchestration**  \n    Go Center handles workflow orchestration (Temporal), project management, agent registry. Rust Edge runs local LLM execution with Docker sandbox. VS Code Plugin provides real-time developer awareness. No single point of failure.\n\n---\n\n## 🗺️ Roadmap\n\n**Core OS** (ongoing):\n- Enhanced MCP tool ecosystem and dynamic discovery\n- Multi-model routing optimization with cost-aware scheduling\n- Knowledge graph query performance and scale improvements\n- Template engine with versioned prompt inheritance\n- Rich event system with fine-grained subscription filters\n\n**Application Layer** (upcoming):\n- **Q3 2026**: Native web dashboard for agent monitoring and task management; Python\u002FTypeScript SDK for easier integration\n- **Q4 2026**: Kubernetes deployment operator; Multi-turn conversation memory compression; Skill marketplace prototype\n- **2027**: Distributed agent mesh across Edge nodes; Multi-modal agent support (vision, audio); Community plugin registry\n\n---\n\n## 📊 Performance Goals\n\n| Operation | Latency | Throughput |\n|-----------|---------|-----------|\n| L2 Node Write (Oxigraph) | ~2ms | 500 ops\u002Fsec |\n| L3 SPARQL Projection | ~15ms | 66 ops\u002Fsec |\n| L0 Sled KV Read | ~1ms | 1000 ops\u002Fsec |\n| Agent ReAct Turn | 1-5s | 0.2-1 turns\u002Fsec |\n| **Idle Memory** | ~200MB | scales with tasks |\n\n---\n\n## 📚 Documentation\n\n- **Design Detail** → [`docs\u002FDESIGN_DETAIL.md`](docs\u002FDESIGN_DETAIL.md) · [`docs\u002FDESIGN_DETAIL.zh.md`](docs\u002FDESIGN_DETAIL.zh.md) (中文)\n- **Core Design Philosophy** → [`docs\u002FCORE_DESIGN_PHILOSOPHY.md`](docs\u002FCORE_DESIGN_PHILOSOPHY.md) · [`docs\u002FCORE_DESIGN_PHILOSOPHY.zh.md`](docs\u002FCORE_DESIGN_PHILOSOPHY.zh.md) (中文)\n- **gRPC Proto** → [`proto\u002Fpdca_core.proto`](proto\u002Fpdca_core.proto)\n\n---\n\n## 🤝 Contributing\n\nWe welcome contributions from the community!\n\n- **🐛 Report bugs**: [GitHub Issues](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse\u002Fissues)\n- **💡 Propose ideas**: [GitHub Discussions](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse\u002Fdiscussions)\n- **🔀 Submit PRs**: Fork → feature branch → PR against `main`\n\n```bash\ngit checkout -b feat\u002Fmy-feature\n# Make your changes\ncargo fmt && cargo clippy  # Keep code clean\ncargo test                 # Ensure nothing breaks\ngit commit -am 'Add my feature'\ngit push origin feat\u002Fmy-feature\n```\n\nAll contributors are expected to adhere to our [Code of Conduct](docs\u002FCODE_OF_CONDUCT.md).\n\n---\n\n## 📄 License\n\nMIT License — see [LICENSE](LICENSE).\n\n---\n\n\u003Cdiv align=\"center\">\n\nStar ⭐ if you find this useful — join us in building the infrastructure for tomorrow's AI.\n\n[![GitHub stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdoiito\u002Fgliding_horse.svg?style=social&label=Star)](https:\u002F\u002Fgithub.com\u002Fdoiito\u002Fgliding_horse)\n\n*\"Wisdom is not inherited; it is built upon the shoulders of those who came before.\"*\n\n\u003C\u002Fdiv>\n",2,"2026-06-11 04:10:51","CREATED_QUERY"]