KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#dsh #dsh-plugin #typescript #developer-tools #claude-code #ai-agent #codex #deepseek-harness #ai-agents #cli #react #mcp

gliding_horse

★ 130 Open GitHub ↗

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.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Gliding Horse Agent OS 
<div align="center">

![Gliding Horse Logo](https://raw.githubusercontent.com/doiito/gliding_horse/main/assets/logo.jpg)

**An Industrial-Grade AI Agent Operating System Built in Rust**  [![Star on GitHub](https://img.shields.io/github/stars/doiito/gliding_horse?style=flat)](https://github.com/doiito/gliding_horse)

*Inspired by Zhuge Liang's Wooden Ox and Gliding Horse — Ancient Ingenuity Meets Modern AI*

[![Rust](https://img.shields.io/badge/Rust-2021-orange.svg)](https://www.rust-lang.org/)
[![License](https://img.shields.io/badge/license-MIT-blue.svg)](https://raw.githubusercontent.com/doiito/gliding_horse/main/LICENSE)
[![gRPC](https://img.shields.io/badge/gRPC-Protocol-green.svg)](https://grpc.io/)
[![Knowledge Graph](https://img.shields.io/badge/Knowledge%20Graph-Oxigraph-purple.svg)](https://oxigraph.org/)
[![Release](https://img.shields.io/badge/release-v0.1.4-blue)](https://github.com/doiito/gliding_horse/releases)

---

[**English**](README.md) · [**中文**](README.zh.md) · [**Design Detail →**](docs/DESIGN_DETAIL.md)
[**medium URL**](https://medium.com/@doiito-sun)
[**中文稀土掘金**](https://juejin.cn/column/7647868075887165450)
[**中文博客园**](https://www.cnblogs.com/doiito)
[**中文CSDN博客**](https://blog.csdn.net/2604_96270735)
[**B站播客**](https://space.bilibili.com/1547455799/lists)

</div>

---

## 🎉 v0.1.4.preview Release

We are proud to announce the **v0.1.4.preview release** of Gliding Horse Agent OS — a major stability and intelligence release with significant architectural refactoring and 85 file changes (+6134 / −4127 lines).

**Highlights:**

| Area | Description |
|------|-------------|
| **SA Module Monolith Decomposition** | Split the 3408-line `sa/mod.rs` into 8 focused modules (`types`, `planning`, `execution`, `intervention`, `agent`, `process`, `stats`, `actions`) — each with a single responsibility. The largest structural refactor in the codebase. |
| **Unified Timeline System** | New `TimeRange` + `TimelineEntry` framework for cross-subsystem temporal queries. Includes exponential time-decay reranking (`apply_time_decay()`) for memory recall — older entries gracefully decay in relevance. |
| **5W2H Dimension Audit** | Formalized dimension-level audit with `AuditStatus` (Pass/Warning/Fail) and automatic causal chain linkage. Failed dimensions now feed into `CausalEngine` for root-cause analysis — no more black-box PASS/FAIL. |
| **Knowledge Graph Context Injection** | Agent system prompts now auto-inject relevant KG entities before each execution turn, ensuring SA interventions and new knowledge are immediately visible to all agents. |
| **Time-Aware System Prompt** | Agents now receive current time and session context via `SystemPromptRegion::TimeAwareness`, enabling time-sensitive reasoning and checkpoint-consistent recovery. |
| **Hyperspace-Integrated Proactive Perception** | `ProactiveEngine` now uses HyperspaceStore semantic search (with time decay λ=0.5) for experience retrieval — replacing L0 tag-substring matching. Graceful fallback to legacy path. |
| **Causal-Integrated Workspace Monitor** | File events (create/modify/delete) now record `CausalObservation` for root-cause traceability. Pre-task snapshots + objective-aware file inventory injection. |
| **LLRU Cold Archive (Skill Graph)** | `SkillGraphStore` auto-archives cold skills (`last_used_at < cutoff`) to L0 with `storage_tier = L0Permanent`. `find_stale_skills()` triggers re-indexing of outdated entries. |
| **TimelineStore Mutation Tracking** | Every `SkillGraphStore` structural change (register/update/remove/link/MOC) now records a `GraphMutation` — making `pending_mutations()` and `snapshot_count()` accurately reflect real graph activity. |
| **Old DAG Workflow Engine Removed** | Removed the 211-line petgraph-based `DagEngine` — PDCA's 7-level adaptive execution now fully replaces the legacy DAG orchestration. |
| **HNSW Lock-Free Safety** | `visited_gen` changed from `Vec<usize>` to `Vec<AtomicUsize>` in `IncrementalHNSW`, eliminating concurrent search data races while preserving lock-free throughput. |
| **PDCA P0: Pre-check Runtime Error** | Fixed crash (`Error("expect L0-3")`) when TL fails to match a skill — gracefully falls back to L0 execution instead of aborting the workflow. |
| **PDCA P1: PA Uncreatable** | Fixed PA creation failure when DA/TL sets `PauseOnError` — missing `execute` field now properly populated. |
| **PDCA P2: No Output on L0 Fallback** | Fixed silent output loss when metrics unavailable — default metrics ensure user always receives a response. |
| **TL: pend always 0** | Fixed `pend_sum` calculation in TL aggregation — was incorrectly using `sum` instead of actual pend values from sub-tasks. |
| **Error Handling Cleanup** | Removed two `.expect("RwLock poisoned")` calls in `execution.rs` — graceful degradation over panic. |
| **Skill Graph Security Enhancements** | New access-control checkpoints in skill registration and query paths, plus MCP tool call security filtering. |

---

## 🎉 v0.1.3 Release

We are proud to announce the **v0.1.3 release** of Gliding Horse Agent OS.

**What's new in v0.1.3:**

| Feature | Description |
|---------|-------------|
| **Causal Engine** | New standalone causal analysis subsystem with `CausalEngine`, `FusionEngine`, `CausalStore`, and typed `CausalFactor`. Enables causal reasoning across agent operations with fused multi-factor analysis — identifies root causes, propagates failure chains, and computes causal graphs for agent decisions. |
| **Unified Graph Backend** | Consolidated `GraphBackend` (~1,200 LOC) replacing fragmented graph storage — provides a single, optimized interface for node/edge CRUD with batch operations, subgraph extraction, and path-finding across all knowledge layers. |
| **Graph Features Computation** | New `graph_features` module computing structural feature vectors (degree centrality, clustering coefficient, PageRank, betweenness) and graph similarity scoring via feature-distance comparison. Enables quantitative graph analysis and comparison across cognitive snapshots. |
| **Snapshot Timeline** | Skill-graph snapshots with durable post-snapshot mutation records, point-in-time restoration, and diff support. It is an experimental graph timeline, not a complete session-history or crash-recovery system. |
| **Self-Awareness (SA) Overhaul** | Major rewrite of the self-awareness module (+410 lines) with enhanced monitoring of agent state, environment perception, and adaptive behavior. Integrated with the causal engine for self-diagnosis. |
| **5W2H Dimension Audit Enhancement** | Expanded dimension-level audit in `src/core/five_w2h.rs` with deeper causal attribution per dimension. What/Why failures now chain into the causal engine for automated root-cause analysis. |
| **Advanced Features Design** | Comprehensive [`ADVANCED_FEATURES_DESIGN.md`](docs/ADVANCED_FEATURES_DESIGN.md) document covering graph backend architecture, causal reasoning design, timeline-based snapshot semantics, and performance benchmarks. |
| **Graph Backend Benchmarks** | New benchmark suite (`benches/bench_graph_backend.rs`) measuring node/edge R/W, subgraph extraction, and path-finding throughput. |
| **Gliding Code TUI Refinements** | Engine and TUI improvements in the terminal client — better Markdown rendering, enhanced MCP server lifecycle, and internal refactoring for maintainability. |
| **Bug Fixes** | Fixed duplicate secondary-index update in L2 `write_node` that caused inconsistent indexing under concurrent writes. |

---

## 🎉 v0.1.2 Release

We are proud to announce the **v0.1.2 release** of Gliding Horse Agent OS.

**What's new in v0.1.2:**

| Feature | Description |
|---------|-------------|
| **HyperspaceEngine** | Production-grade vector embedding engine with HNSW ANN search, Write-Ahead Log (WAL), tangent-space pruning, and runtime-switchable metrics (Poincaré, Cosine, Euclidean, Lorentz). |
| **Skill Graph Cognitive Network** | Hypergraph composition, Poincaré structural embeddings, PageRank/betweenness/community detection algorithms, causal failure analysis, experimental temporal snapshots/rollback, formal invariant verification (6 checks), hybrid text×structural search. |
| **Semantic Skill Discovery** | Vector-store integration in `SkillDiscoveryEngine` — finds semantically related skills via HyperspaceStore cosine search, replaces Jaccard-only `suggest_links()` |
| **Oxigraph SPARQL Bridge** | Skill-graph-to-Oxigraph RDF projection via SPARQL INSERT/DELETE with named-graph isolation; reverse RDF-to-skill synchronization is not implemented. |
| **L2 Blackboard Memory** | Typed document store with JSON-LD threading, projections, message packs, and LRU eviction for long-term agent context |
| **Workspace Monitor** | Real-time file system perception engine with 10 event triggers, anomaly deduplication, and 5W2H constraint checking |
| **Batch Agent Manager** | Sliding-window batch components with configurable triggers, event-bus integration, and business-domain isolation. The root gRPC service wires opt-in custom-event and cron/window consumption, including streaming requests through shared service state; its L0 event journal deduplicates, acknowledges on successful execution, and replays pending input at startup. `KnowledgePersister` is not wired to this execution path; graph-changing handlers are disabled unless `apply_graph_mutations` is explicitly enabled, and that opt-in path still lacks event-id idempotency and approval/transaction gates. |
| **Gliding Code TUI** | Interactive terminal UI (ratatui v0.28) with Markdown rendering, MCP server support, checkpoint/resume, and multi-model backends |

---

## What Is Gliding Horse?

An **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.

> "We don't just build agents; we build the **infrastructure that harnesses their collective intelligence**."

### Core Architecture

| Layer | Technology | Role |
|-------|-----------|------|
| **Core Coordination** (Rust) | `PDCA cycle` · `5W2H ontology` · `EventBus` | Agent orchestration & lifecycle |
| **Skill Graph** | `RDF` · `6 link types` · `18 modules` | Dynamic cognitive network |
| **Memory System** | `L0 Sled` · `L1 Session` · `L2 Blackboard` · `L3 Projection` · `MESI coherence` | Hierarchical memory with prefetch |
| **Knowledge Graph** | `Oxigraph RDF` · `SPARQL 1.1` · `Code AST` · `Named Graphs` | Cross-subsystem unified store |
| **HyperspaceEngine** | `HNSW ANN` · `WAL` · `Poincaré/Cosine/Euclidean` · `Hybrid search` | Embedded vector embeddings |
| **Gliding Code TUI** | `ratatui` · `crossterm` · `MCP` · `checkpoint/resume` | Terminal AI coding assistant |
| **Data Bus** | `JSON-LD 1.1` · `@id/@type/@context` · `Named Graphs` | Universal interoperability |
| **Gateway** | `gRPC` · `HTTP (OpenAI-compatible)` · `MCP` | Production interface |
| **Perception Engine** | `10 triggers` · `Anomaly dedup` · `5W2H constraint check` | Proactive monitoring |
| **Agent Workflow** | `PA/DA/CA` · `Tool system` · `Checkpoint` · `Tracked actions` | Multi-agent execution |

---

## 📖 The Story: From Ancient Wisdom to Modern Intelligence

In 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.

His 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**.

### Bridging Past and Present

Just 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**:

| Ancient Innovation | Modern Implementation |
|-------------------|----------------------|
| **Autonomous Transport** | Self-directing agent workflows |
| **Terrain Adaptation** | Dynamic complexity handling (7 levels) |
| **Load Distribution** | Parallel agent execution |
| **Minimal Guidance** | Proactive anomaly detection |
| **Mechanical Reliability** | Rust's memory safety guarantees |

> *"The wise adapt their methods to circumstances, just as water shapes its course according to the ground over which it flows."*  
> — **Zhuge Liang**

This ancient wisdom guides our design: **flexible orchestration that adapts to task complexity**, rather than rigid frameworks that force tasks into predefined molds.

---

## 🔧 Key Highlights

### 1. HyperspaceEngine — Embedded Vector Engine
Production-grade spatial memory engine with **runtime-switchable metrics** (Poincaré, Cosine, Euclidean, Lorentz). Features **HNSW approximate nearest neighbor search**, CRC32-verified **Write-Ahead Log (WAL)** with 3 sync modes, **tangent-space pruning** for Poincaré ball search, JSON-LD metadata index with RoaringBitmap filters, and dual-space **hybrid search** (text × structural). A self-contained crate with zero external vector database dependencies.

### 2. Skill Graph Cognitive Network
Dynamic in-memory cognitive network with **6 semantic link types** (Prerequisite, Composition, Related, Alternative, Extends, Generalization). Includes **Poincaré structural embedding** computation from graph topology (prerequisite depth, tag fingerprinting), **hypergraph composition** with first-class `Hyperedge` and `CompositionType` (Sequential, Parallel, Conditional, Optional, Fallback), **graph algorithms** (PageRank, betweenness centrality, label-propagation community detection, DFS prerequisite chains, Tarjan SCC cycle detection), **causal failure analysis** with root cause inference, **formal invariant verification** (6 checks: acyclicity, link existence, composite reachability, no deprecated prereqs, valid 5W2H, valid security levels), and **temporal versioning** with snapshot/rollback.

### 3. Generalized PDCA — 7-Level Adaptive Execution
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. **PA/DA/CA agent roles** with template-driven prompt construction.

### 4. CPU Cache-Inspired Memory — 4 Layers + MESI Coherence
**L0** Sled disk storage → **L1** session context → **L2** Oxigraph RDF + Blackboard → **L3** SPARQL projection cache. The repository implements cache-inspired coordination and prefetch components; no end-to-end latency or multi-agent consistency benchmark is currently published.

### 5. JSON-LD Universal Data Bus — Internal Interoperability Subset
The internal JSON-LD utilities support `@context`, `@id`, `@graph`, framing, validation and routing used by this repository. They are not a claim of complete JSON-LD 1.1, SHACL, or general RDF interoperability.

### 6. Self-Evolving Skill Graph — Autonomous Learning
AA agents record knowledge fragments, links, and evolution suggestions after task completion. `/learn` and `/reduce` provide explicit acquisition/consolidation operations; suggestions are not automatically applied because typed-patch verification, security, and conflict gates are still pending. `BootstrapEngine` ingests markdown skills from the filesystem.

### 7. Universal Knowledge Graph — Unified Cognitive Backbone
Skills, memories, tasks, and code knowledge can use the shared **Oxigraph RDF store** through named graphs, enabling scoped SPARQL joins where producers are wired to that store. Code ASTs parsed by tree-sitter are converted to RDF triples. `SkillGraphStore` projects its changes into the semantic store; reverse RDF-to-skill synchronization is not implemented.

### 8. Semantic Skill Discovery Engine
`SkillDiscoveryEngine` wraps `HyperspaceStore` for vector-based semantic search across skills. `suggest_links()` falls back from Jaccard tag overlap to cosine similarity via embedding vectors. Includes BFS path finding (`find_skill_chain()`), composition tree construction (`get_skill_tree()`), and conflict detection.

### 9. 5W2H Dimension-Level Audit — Precision Rollback
CA audits each of the 7 dimensions independently. What/Why fail → re-analyze. How/Where fail → re-plan. When/HowMuch fail → conditional pass. No more black-box "PASS/FAIL" — you know exactly what went wrong.

### 10. Proactive Perception Engine
10 execution triggers with 60-second anomaly deduplication. Monitors deadline violations, budget overruns (>80% tokens), role mismatches, environment conflicts. **Workspace Monitor** detects file creations/modifications/deletions in real-time. Auto-escalates to human when needed.

### 11. Micro-Tool System — Tame Large Outputs
Results >8KB auto-generate conversational micro-tools (e.g., "search_in_results"). Transforms unwieldy 50KB+ outputs into interactive, queryable artifacts within the LLM context.

### 12. MCP Integration — One Protocol to Connect Them All
Standard **Model Context Protocol** connects GitHub, Slack, Jira, and any MCP-compatible server. Dynamic tool discovery at runtime. Supports both HTTP SSE and stdio transport modes with repeatable `--mcp-server` CLI flags.

### 13. Checkpoint & Recovery — Explicit Session Management
Session checkpoints and `--resume <task_iri>` / `--list-checkpoints` support explicit session management. Crash recovery and complete long-running-task replay require dedicated fault-injection and end-to-end validation before being claimed.

### 14. Center + Edge Federation — Local Autonomy, Global Orchestration
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.

---

## 🖥️ Gliding Code — The Terminal AI Assistant

**Gliding Code** is a terminal-based AI coding assistant (`ratatui` TUI) that brings the power of Gliding Horse's knowledge graph and agent orchestration directly into your command line — no IDE required.

**Features:**
- Interactive TUI with **Markdown rendering** (`tui-markdown`) and **mermaid diagram** support
- **MCP server integration** via `--mcp-server` and `--mcp-server-stdio` flags
- **Checkpoint/resume** with `--resume <task_iri>` and `--list-checkpoints`
- **Multi-model backends**: DeepSeek, OpenAI-compatible APIs
- **PDCA workflow execution** with plan/do/check/act cycles
- **Configurable** workspace, max iterations, max PDCA cycles, verbosity

![Gliding Code Demo](https://raw.githubusercontent.com/doiito/gliding_horse/main/assets/screenshot.gif)

![Knowledge Graph in Action](https://raw.githubusercontent.com/doiito/gliding_horse/main/assets/gliding_code_kg.JPG)
*Knowledge graph visualization — real-time entity relationships, code structure understanding, and cross-subsystem awareness powered by Oxigraph RDF*

![Completed Programming Task](https://raw.githubusercontent.com/doiito/gliding_horse/main/assets/gliding_code.JPG)
*Task completion interface — AI agent successfully analyzing and solving a programming task with full traceability*

---

## 🚀 Quick Start

### Download & Run — Gliding Code

No dependencies required. Just download, extract, and run:

| Platform | Download |
|----------|----------|
| Linux (x86_64, musl) | [`glidingcode-x86_64-unknown-linux-musl.tar.gz`](https://github.com/doiito/gliding_horse/releases) (~15 MB) |
| Linux (aarch64, musl) | [`glidingcode-aarch64-unknown-linux-musl.tar.gz`](https://github.com/doiito/gliding_horse/releases) (~14 MB) |
| macOS (Apple Silicon) | [`glidingcode-aarch64-apple-darwin.tar.gz`](https://github.com/doiito/gliding_horse/releases) (~13 MB) |
| Windows (x86_64) | [`glidingcode-x86_64-pc-windows-msvc.zip`](https://github.com/doiito/gliding_horse/releases) (~12 MB) |

```bash
# Linux / macOS
tar xzf glidingcode-*.tar.gz
./glidingcode --help

# Windows (PowerShell)
Expand-Archive glidingcode-x86_64-pc-windows-msvc.zip .
.\glidingcode.exe --help
```

> All Linux builds are **fully statically linked** (musl) — no runtime dependencies required.

Set your API key and start using it:

```bash
export DEEPSEEK_API_KEY="sk-..."        # Linux / macOS
# or
set DEEPSEEK_API_KEY="sk-..."            # Windows (cmd)
# or
$env:DEEPSEEK_API_KEY="sk-..."           # Windows (PowerShell)

# Alternatively, use any OpenAI-compatible provider:
export AGENT_OS_GATEWAY_API_KEY="sk-..."
export AGENT_OS_GATEWAY_API_URL="https://your-endpoint/v1"

# Web search tool (powered by Exa):
# Get your free API key at https://exa.ai/docs/reference/team-management/get-api-key
# Falls back to DuckDuckGo (unreliable in China, not recommended for Chinese users)
export EXA_API_KEY="your-exa-api-key"

# Run an interactive session
./glidingcode

# Or run a one-shot task
./glidingcode "Explain how Rust's borrow checker works"

# With MCP server attached
./glidingcode --mcp-server chrome=http://localhost:3000/sse

# Resume from checkpoint
./glidingcode --resume task:abc123
```

### Build from Source

```bash
git clone https://github.com/doiito/gliding_horse.git
cd gliding_horse

# Build the glidingcode binary (release, ~51 MB)
cargo build -p code_cli --release
./target/release/glidingcode --help
```

---

## 🗺️ Roadmap

**v0.1.x Release Series** (stabilization):
- Binary distribution for Linux/macOS/Windows via GitHub Releases
- Pre-built musl static builds for Linux (zero-dependency)
- MCP tool ecosystem expansion and documentation
- Checkpoint/resume polish and testing

**v0.2.x Release Series** (planned):
- Native web dashboard for agent monitoring and task management
- Python/TypeScript SDK for easier integration
- Skill marketplace prototype with community plugin registry
- Multi-model routing with cost-aware scheduling

**v0.3.x+ Release Series** (future):
- Kubernetes deployment operator for production scaling
- Distributed agent mesh across Edge nodes
- Multi-modal agent support (vision, audio)
- Multi-turn conversation memory compression

---

## 📊 Performance Targets

| Operation | Latency | Throughput |
|-----------|---------|-----------|
| L2 Node Write (Oxigraph) | ~2ms | 500 ops/sec |
| L3 SPARQL Projection | ~15ms | 66 ops/sec |
| L0 Sled KV Read | ~1ms | 1000 ops/sec |
| Hyperspace HNSW Search (10K vectors) | ~1ms | 1000 qps |
| Poincaré Embedding (4D) | ~50µs | — |
| Agent ReAct Turn | 1-5s | 0.2-1 turns/sec |
| Idle Memory | ~200MB | scales with tasks |

---

## 📚 Documentation

- **Design Detail** → [`docs/DESIGN_DETAIL.md`](docs/DESIGN_DETAIL.md) · [`docs/DESIGN_DETAIL.zh.md`](docs/DESIGN_DETAIL.zh.md) (中文)
- **Core Design Philosophy** → [`docs/CORE_DESIGN_PHILOSOPHY.md`](docs/CORE_DESIGN_PHILOSOPHY.md) · [`docs/CORE_DESIGN_PHILOSOPHY.zh.md`](docs/CORE_DESIGN_PHILOSOPHY.zh.md) (中文)
- **gRPC Proto** → [`proto/pdca_core.proto`](proto/pdca_core.proto)

---

## 🤝 Contributing

We welcome contributions from the community!

- **🐛 Report bugs**: [GitHub Issues](https://github.com/doiito/gliding_horse/issues)
- **💡 Propose ideas**: [GitHub Discussions](https://github.com/doiito/gliding_horse/discussions)
- **🔀 Submit PRs**: Fork → feature branch → PR against `main`

```bash
git checkout -b feat/my-feature
# Make your changes
cargo fmt && cargo clippy  # Keep code clean
cargo test                 # Ensure nothing breaks
git commit -am 'Add my feature'
git push origin feat/my-feature
```

All contributors are expected to adhere to our [Code of Conduct](docs/CODE_OF_CONDUCT.md).

---

## 📄 License

MIT License — see [LICENSE](LICENSE).

---

Star ⭐ if you find this useful — join us in building the infrastructure for tomorrow's AI.

[![GitHub stars](https://img.shields.io/github/stars/doiito/gliding_horse.svg?style=social&label=Star)](https://github.com/doiito/gliding_horse)

*"Wisdom is not inherited; it is built upon the shoulders of those who came before."*

</div>

gliding_horse

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README