Collection · Stated support, not inferred

Runs on Windows, and says so

7 picks · 5 of the atlas's 14 topics · 952,013 combined stars · snapshot 2026-10-22

Most agentic tooling is written on a Mac and tested on Linux, and "it's Python, it'll be fine" is how a Windows afternoon disappears. The atlas records a per-platform verdict for every project, and it distinguishes stated support from support merely inferred from the language. Every project below has documented Windows support -- the build refuses to publish this page if one of them stops.

Open all 7 in the atlas →

From there: save them to your own projects, or export the set as Markdown, a standalone HTML page, or a PDF.

Runs on: Windows WSL2 macOS Linux Docker

  1. 1 The framework

    openclaw

    openclaw/openclaw

    The most-starred project in the atlas, and it puts Windows in the same sentence as macOS and Linux rather than in a footnote. Three source lists carry it.

    Open-source AI agent framework that turns LLMs into persistent, proactive personal AI agents with multi-channel messaging (Signal, Telegram, Discord, WhatsApp), cron scheduling, memory systems, MCP integration, skill plugins, sub-agent spawning, and browser automation

    npm install -g openclaw@latest --allow-scripts=openclaw

    392,696 stars · 6 of 39 lists · TypeScript · MIT · pushed 2026-10-22

    Windows WSL2 macOS Linux Docker

    Frameworks & SDKs MCP

  2. 2 The harness

    Pi

    earendil-works/pi

    A deliberately small terminal harness built around skills loaded on demand. Small is what makes it portable: there is less of it to be Unix-shaped.

    Minimal terminal coding harness built around "lazy skills": each capability keeps only a one-line description in active context, loading full instructions and tool schemas only when invoked. Keeps the system prompt under 1,000 tokens versus 7,000–10,000 for typical agents, making it a concrete reference for context-minimal harness design

    npm install -g --ignore-scripts @earendil-works/pi-coding-agent

    116,957 stars · 1 list · TypeScript · MIT · pushed 2026-10-21

    Windows WSL2 macOS Linux Docker

    Harnesses & Runtime Infra

  3. 3 The coding agent

    Continue

    continuedev/continue

    Ships as an editor extension, which is why it works on Windows without a WSL detour -- the editor has already solved that problem.

    Continue is an open-source autopilot plugin for VS Code and JetBrains, enhancing coding with LLMs through features like task and tab autocomplete, natural language edits, file generation, and customization options, available under the Apache 2.0 license github | website

    git clone https://github.com/continuedev/continue.git

    36,290 stars · 7 of 39 lists · TypeScript · Apache-2.0 · pushed 2026-10-20

    Windows WSL2 macOS Linux Docker

    Coding Agents MCPCodex / OpenAI

  4. 4 The model runner

    Ollama

    ollama/ollama

    A native Windows build, a one-line pull, and no Python environment to get wrong. If you want a local model on Windows this is the shortest path to one.

    Ollama is a tool for running large language models locally, offering easy setup for macOS, Windows, Linux, and Docker, along with a library of models and quickstart guides for customization and integration github | github profile

    irm https://ollama.com/install.ps1 | iex

    184,208 stars · 4 of 39 lists · Go · MIT · pushed 2026-10-22

    Windows WSL2 macOS Linux Docker

    Harnesses & Runtime Infra Local / Ollama

  5. 5 The inference engine

    llama.cpp

    ggml-org/llama.cpp

    The C++ layer under most of local AI, with first-class Windows builds and no runtime to install. Worth knowing directly once the layer above it surprises you.

    C/C++ inference. CPU, GPU, Apple Silicon. Foundation of local AI

    irm https://llama.app/install.ps1 | iex

    131,944 stars · 3 of 39 lists · C++ · MIT · pushed 2026-10-21

    Windows WSL2 macOS Linux Docker

    Harnesses & Runtime Infra Local / Ollama

  6. 6 The context budget

    headroom

    headroomlabs-ai/headroom

    Compresses tool output, logs and files before they reach the window. Windows and Docker support are both stated, and it sits in front of whatever agent you chose.

    Compresses tool outputs, logs, files, and RAG chunks before they enter the context window, cutting active tokens by 60–95% without changing answers. Ships as a library, proxy, and MCP server — the right drop-in layer for any harness where bulky tool returns are the primary context pressure source

    uv tool install --python 3.13 "headroom-ai[all]" # CLI in a self-contained env

    77,253 stars · 2 of 39 lists · Python · Apache-2.0 · pushed 2026-10-22

    Windows WSL2 macOS Linux Docker

    Plugins, Themes & Clients Claude CodeClaude / AnthropicMCPLangChain / LangGraph

  7. 7 The sandbox

    ironclaw

    nearai/ironclaw

    Treats agent execution as a privacy and isolation problem rather than a convenience one. Two source lists agree, and Windows support is documented rather than assumed.

    NEAR AI's open-source agent OS that treats agent execution as a privacy-first harness problem: untrusted tools run in WASM sandboxes with capability-based permissions, credentials are injected at the host boundary with leak detection, and prompt-injection filtering plus endpoint allowlisting constrain agent behavior. The combination of local encrypted storage, hybrid-search memory, MCP server supp

    irm "https://github.com/nearai/ironclaw/releases/download/$IronClawReleaseTag/ironclaw-installer.ps1" | iex

    12,665 stars · 3 of 39 lists · Rust · Apache-2.0 · pushed 2026-10-20

    Windows WSL2 macOS Linux Docker

    Assistants & Domain Agents MCP

The whole set, in one paste

Every install command above, in the order they are listed. Read them before you run them — they are other people's one-liners, and 3 of them pipe a download straight into a shell.

npm install -g openclaw@latest --allow-scripts=openclaw
npm install -g --ignore-scripts @earendil-works/pi-coding-agent
git clone https://github.com/continuedev/continue.git
irm https://ollama.com/install.ps1 | iex
irm https://llama.app/install.ps1 | iex
uv tool install --python 3.13 "headroom-ai[all]" # CLI in a self-contained env
irm "https://github.com/nearai/ironclaw/releases/download/$IronClawReleaseTag/ironclaw-installer.ps1" | iex