[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-42":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":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":27,"readmeContent":28,"aiSummary":29,"trendingCount":16,"starSnapshotCount":16,"syncStatus":30,"lastSyncTime":31,"discoverSource":32},42,"OpenMonoAgent.ai","StartupHakk\u002FOpenMonoAgent.ai","StartupHakk","(BETA) AI shouldn't have a meter. Unlimited tokens. Forever. Your machine. Your agent. Use it from anywhere. Terminal-native coding agent powered by local LLMs — 100% open source, free forever, and installed with a single command. Proudly built on C#\u002F.NET, because AI tooling should be infrastructure, not a subscription.","",null,"C#",1530,188,35,17,0,20,104,674,60,19.83,"Other",false,"main",true,[],"2026-06-12 02:00:07","\u003Cdiv align=\"center\">\n  \u003Cimg src=\"docs\u002Fassets\u002Flogo.png\" alt=\"OpenMonoAgent\" width=\"480\" \u002F>\n\u003C\u002Fdiv>\n\n\u003Cdiv align=\"center\">\n  \u003Cstrong>Open-source coding agent. Local-first. Zero cost. Zero cloud.\u003C\u002Fstrong>\u003Cbr\u002F>\n  \u003Csub>Built to democratize AI. Powered by .NET.\u003C\u002Fsub>\n\u003C\u002Fdiv>\n\n\u003Cbr>\n\n\u003Cdiv align=\"center\">\n  \u003Ca href=\"#quickstart\">Quickstart\u003C\u002Fa> · \u003Ca href=\"#how-it-compares\">How it compares\u003C\u002Fa> · \u003Ca href=\"#whats-inside\">What's inside\u003C\u002Fa> · \u003Ca href=\"#supported-hardware\">Hardware\u003C\u002Fa> · \u003Ca href=\"#docs\">Docs\u003C\u002Fa> · \u003Ca href=\"ROADMAP.md\">Roadmap\u003C\u002Fa> · \u003Ca href=\"#contributing\">Contributing\u003C\u002Fa>\n\u003C\u002Fdiv>\n\n\u003Cbr>\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fstatus-beta-FF8C00?style=for-the-badge&labelColor=555555\" alt=\"Status: Beta\" \u002F>\n\u003C\u002Fdiv>\n\n\u003Cbr>\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F.NET-10-512BD4?logo=dotnet&logoColor=white\" alt=\".NET 10\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-AGPL--3.0-green\" alt=\"GNU AGPL-3.0 License\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fdocker-ready-2496ED?logo=docker&logoColor=white\" alt=\"Docker\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fllama.cpp-local%20inference-black?logo=llama&logoColor=white\" alt=\"llama.cpp\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fself--hosted-yes-brightgreen\" alt=\"Self-hosted\" \u002F>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fplatform-Linux-FCC624?logo=linux&logoColor=black\" alt=\"Linux\" \u002F>\n\u003C\u002Fdiv>\n\n---\n\nOpenMono is a coding agent that runs entirely on your hardware — no subscriptions, no data leaving your network, no per-token billing. It pairs a .NET 10 CLI with its own llama.cpp inference server, giving you a full agentic loop with 20 built-in tools, Docker sandboxing, and deep code intelligence. GPU or CPU, it auto-configures itself. You own the model, the compute, and the data.\n\n---\n\n## Quickstart\n\n```\nbash \u003C(curl -fsSL https:\u002F\u002Fraw.githubusercontent.com\u002FStartupHakk\u002FOpenMonoAgent.ai\u002Frefs\u002Fheads\u002Fmain\u002Fget-openmono.sh)\n```\n\nThen from any project:\n\n```bash\ncd your-project\u002F\n\nopenmono agent          # TUI mode (default)\nopenmono agent --classic    # classic scrolling terminal\n```\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"docs\u002Fassets\u002Ftui-snapshot-openmono.png\" alt=\"OpenMono TUI\" width=\"780\" \u002F>\n\u003C\u002Fdiv>\n\n> [!NOTE]\n> TUI mode is the default for interactive terminals. Use `openmono  agent --classic` for CLI.\n\n→ [Full command reference](docs\u002FSETUP.md) — daily commands, setup flags, GPU\u002FCPU options\n\n---\n\n## How it compares\n\nMost coding agents are cloud products wearing an open-source label. Your prompts, your code, and your context hit someone else's servers on every keystroke. You pay per token, forever, with no ceiling.\n\nOpenMono runs the model on your hardware via llama.cpp — an RTX 3090 or a workstation NUC is all you need. After the one-time setup, inference costs nothing. Your code never leaves the machine. No account, no usage dashboard, no API key.\n\nIt's a full [agentic loop](docs\u002FARCHITECTURE.md): 20 tools, sub-agents, Docker sandboxing, LSP code intelligence, native Roslyn C# analysis, MCP integration, and [playbooks](docs\u002FPLAYBOOKS.md). Runs at ~45 tok\u002Fs on GPU, ~20 tok\u002Fs on CPU.\n\n\n|  | **OpenMono** | Claude Code | OpenCode |\n|--|:-------------|-------------|----------|\n| **Inference cost** | Zero per token (local) | Per-token billing | Per-token billing |\n| **Data privacy** | Fully offline capable | Cloud only | Depends on provider |\n| **Default inference** | llama.cpp bundled, zero config | Anthropic API required | BYO provider, no bundled inference |\n| **Sandboxing** | Docker-native | Host process | Host process |\n| **Code intelligence** | LSP + Roslyn + MCP graph tools | File reads | LSP (30+ servers) |\n| **Extensibility** | [Playbooks](docs\u002FPLAYBOOKS.md) (typed, composable) | Skills (markdown) | Plugins (TS SDK) |\n| **MCP** | Client (stdio) | Full client | Full client |\n| **UI** | TUI + CLI | Web, Desktop, VS Code, CLI | TUI, Desktop, Web |\n---\n\n## What's inside\n\n\u003Ctable>\n\u003Ctr>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n**01 · Bundled inference — zero config, zero cost**  \nllama.cpp ships inside Docker. Installer detects your hardware and picks the right model. After setup, every token is free.\n\n`GPU` Qwen3.6-27B dense · ~60 tok\u002Fs  \n`CPU` Qwen3.6-35B-A3B MoE · ~20 tok\u002Fs\n\n→ [Models & reasoning mode](docs\u002FMODELS.md)\n\n\u003C\u002Ftd>\n\u003Ctd width=\"50%\" valign=\"top\">\n\n**02 · Agentic loop that earns its name**  \n25 iterations per turn. Doom-loop detection aborts if the same tool sequence repeats 3×. Checkpoints at 65% context fill, compacts at 80%. Runs until done — then stops.\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd valign=\"top\">\n\n**03 · [20 tools](docs\u002FARCHITECTURE.md), 12-step pipeline**  \nEvery call: parse → schema validate → path sanity → plan-mode guard → capability check → cache → pre-hook → execute → post-hook → artifact store. Read-only tools run in parallel. Nothing bypasses the pipeline.\n\n\u003C\u002Ftd>\n\u003Ctd valign=\"top\">\n\n**04 · 5 specialist sub-agents**  \nIsolated sessions with locked tool sets and turn budgets:\n\n`Explore` · read-only discovery · 15 turns  \n`Plan` · architecture, no writes · 10 turns  \n`Coder` · full file access · 30 turns  \n`Verify` · adversarial + Roslyn · 20 turns  \n`general-purpose` · everything · 25 turns\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd valign=\"top\">\n\n**05 · Docker sandbox**  \nProject mounts as `\u002Fworkspace`. The agent reads and writes your real files — that's the blast radius. Nothing outside that mount is visible or reachable.\n\n\u003C\u002Ftd>\n\u003Ctd valign=\"top\">\n\n**06 · Deep code intelligence**  \nRoslyn: type hierarchy, blast-radius, cross-file symbol search, callers, diagnostics — 5-min compilation cache. LSP for TypeScript, Python, Go, Rust, lazy-started on first use.\n\nAuto-detects [graphify](docs\u002Fgraphify.md) (semantic concept graph, 25+ languages) and [code-review-graph](docs\u002Fcode-review-graph.md) (structural call graph via MCP, ~22 tools) if installed — no config needed.\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003Ctr>\n\u003Ctd valign=\"top\">\n\n**07 · [Playbooks](docs\u002FPLAYBOOKS.md)**  \nYAML workflows with typed parameters, conditional gates, and checkpoint\u002Fresume. Composable — one playbook can call another.\n\n\u003C\u002Ftd>\n\u003Ctd valign=\"top\">\n\n**08 · [4 providers](docs\u002FMODELS.md), hot-swappable**  \nLocal llama.cpp is the default and fully supported. OpenAI, Anthropic, and Ollama are available but WIP — see [Models](docs\u002FMODELS.md) for details.\n\n\u003C\u002Ftd>\n\u003C\u002Ftr>\n\u003C\u002Ftable>\n\n> **09 · Distributed inference** — Agent on your laptop, inference on a separate GPU machine. Free relay at [app.openmonoagent.ai](https:\u002F\u002Fapp.openmonoagent.ai).\n\n\u003Cdiv align=\"center\">\n  \u003Cimg src=\"docs\u002Fassets\u002Fdual-box-server.png\" alt=\"Distributed inference: agent on laptop, inference on GPU machine\" width=\"680\" \u002F>\n\u003C\u002Fdiv>\n\n---\n\n## Supported Hardware\n\n| | GPU | CPU |\n|--|-----|-----|\n| **Model** | Qwen3.6-27B-Q4_K_M | Qwen3.6-35B-A3B-UD-Q4_K_XL |\n| **Minimum** | 24 GB VRAM | 24 GB RAM |\n| **Speed** | ~45–50 tok\u002Fs (RTX 3090) | ~17–20 tok\u002Fs |\n| **Auto-configured** | ✓ | ✓ |\n\n> [!NOTE]\n> The installer detects your hardware and selects the right model automatically — no config needed. Requires Ubuntu 26.04 LTS (recommended) or 25.10.\n\n## Architecture\n\nA .NET 10 CLI driving a local llama.cpp inference server over HTTP, everything sandboxed in Docker. The agent streams tokens, dispatches tool calls through a 12-step pipeline, and loops until done.\n\n→ [Full architecture + diagram](docs\u002FARCHITECTURE.md)\n\n## Configuration\n\nSettings load from `~\u002F.openmono\u002Fsettings.json` (user-level) or `.openmono\u002Fsettings.json` (project-level) — reference, providers, permissions, MCP servers\n\n→ [Full configuration reference](docs\u002FCONFIG.md)\n\n## Commands & shortcuts\n\n14 slash commands including `\u002Fthink`, `\u002Fundo`, `\u002Fresume`, and `\u002Fexport`. Full keyboard shortcut reference for TUI mode.\n\n→ [Commands, slash commands & keyboard shortcuts](docs\u002FSETUP.md)\n\n## Docs\n\n- [Roadmap](ROADMAP.md)\n- [Setup & commands](docs\u002FSETUP.md) — daily commands, TUI vs classic, flags\n- [Architecture](docs\u002FARCHITECTURE.md) — .NET CLI + llama.cpp + Docker, full diagram\n- [Models & reasoning mode](docs\u002FMODELS.md)\n- [Configuration](docs\u002FCONFIG.md) — settings.json, providers, permissions, MCP servers\n- [Tools](docs\u002FARCHITECTURE.md)\n- [Playbooks](docs\u002FPLAYBOOKS.md)\n- [graphify](docs\u002Fgraphify.md) — semantic code graph, 25+ languages\n- [code-review-graph](docs\u002Fcode-review-graph.md) — structural call graph via MCP\n- [Contributing](CONTRIBUTING.md)\n\n---\n## WIP\n\n- The agent goes to max 25 iterations at the moment which is configured in [ConversationLoop.cs Line 149](https:\u002F\u002Fgithub.com\u002FStartupHakk\u002FOpenMonoAgent.ai\u002Fblob\u002F9b6e95b037c163cfab017f244b09e104f1e7e6a8\u002Fsrc\u002FOpenMono.Cli\u002FSession\u002FConversationLoop.cs#L149)\nThe team is currently working on a fix for it, but increasing this number to a very big value may lead to delayed responses and context window getting filled quickly.\n---\n\n> [!NOTE]\n> OpenMono is in **Public Beta**. Early access is open, and we're shipping updates fast. Try it out and tell us what you'd like to see next.\n\n## Contributing\n\nOpenMono is early and moving fast. Contributions are welcome — new tools, providers, LSP servers, playbooks, bug fixes, or docs.\n\nRead the [contributing guide](CONTRIBUTING.md) before opening a PR.\n\n---\n\n\u003Cdiv align=\"center\">\n  \u003Cbr>\n  \u003Cem>\"AI shouldn't be a subscription you rent. It should be infrastructure you own —\u003Cbr>sitting on your desk, serving your code, answering only to you.\"\u003C\u002Fem>\u003Cbr>\u003Cbr>\n  \u003Csub>— Startup Hakk\u003C\u002Fsub>\n\u003C\u002Fdiv>\n\n\u003Cbr>\n\n\u003Cdiv align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fstartuphakk.com\">\u003Cimg src=\"docs\u002Fassets\u002FSTARTUP-HAKK-logo.jpg\" alt=\"StartupHakk\" width=\"140\" \u002F>\u003C\u002Fa>\u003Cbr>\n  \u003Csub>GNU AFFERO GENERAL PUBLIC LICENSE v3.0 · © 2026 StartupHakk\u003C\u002Fsub>\n\u003C\u002Fdiv>\n","OpenMonoAgent.ai 是一个基于本地运行的开源编码助手，旨在通过本地硬件提供无限令牌的AI支持，而无需任何云服务或订阅费用。该项目采用C#\u002F.NET构建，结合了llama.cpp推理服务器，能够自动配置以适应用户的GPU或CPU环境。它内置了20种工具，并支持Docker沙箱和深度代码智能分析，确保用户的数据完全私有化处理。适用于追求数据隐私保护、偏好本地计算资源管理以及希望避免长期订阅成本的开发者或团队使用。",2,"2026-06-11 02:30:36","CREATED_QUERY"]