docs: rebrand to loopforge in docs and readme

Author: rexleimo

README.md

@@ -1,8 +1,14 @@
-# RexOS
+# LoopForge
 
 English | [简体中文](README.zh-CN.md)
 
-RexOS is a long-running agent operating system: persistent memory, tool sandboxing, and model routing, plus an Anthropic-style harness for multi-session work.
+LoopForge (formerly RexOS) is a long-running agent operating system: persistent memory, tool sandboxing, and model routing, plus an Anthropic-style harness for multi-session work.
+
+## Brand update
+
+- Public product name: **LoopForge**
+- Compatibility names still in use: `rexos` (CLI), `~/.rexos` (config/data dir), and `rexleimo/rexos` (repo path)
+- Existing scripts/docs using `rexos` continue to work
 
 ## Documentation
 
@@ -39,13 +45,13 @@ cargo build --release -p rexos-cli
 
 ## Run with Ollama (OpenAI-compatible)
 
-RexOS defaults to `ollama` at `http://127.0.0.1:11434/v1` in `~/.rexos/config.toml`.
+LoopForge defaults to `ollama` at `http://127.0.0.1:11434/v1` in `~/.rexos/config.toml`.
 
 ```bash
 # 1) Start Ollama
 ollama serve
 
-# 2) Init RexOS (creates ~/.rexos/config.toml + ~/.rexos/rexos.db)
+# 2) Init LoopForge (compat command: rexos)
 rexos init
 
 # 3) Run an agent session in a workspace directory
@@ -69,7 +75,7 @@ git push origin v0.1.0
 
 ## Providers & routing
 
-RexOS supports multiple LLM providers via drivers:
+LoopForge supports multiple LLM providers via drivers:
 - `openai_compatible` (Ollama / DeepSeek / Kimi / Qwen / GLM / MiniMax / NVIDIA NIM / OpenAI-compatible gateways)
 - `dashscope_native` (Alibaba DashScope Generation API / Qwen native)
 - `zhipu_native` (Zhipu GLM native auth/token handling)
@@ -97,7 +103,7 @@ provider = "ollama"
 model = "default" # uses providers.<name>.default_model
 ```
 
-To switch providers, set the provider's `api_key_env` (if needed) and update `[router.*]` to point at the provider you want. If you keep `model = "default"`, RexOS uses `providers.<name>.default_model`.
+To switch providers, set the provider's `api_key_env` (if needed) and update `[router.*]` to point at the provider you want. If you keep `model = "default"`, LoopForge uses `providers.<name>.default_model`.
 
 Built-in presets include:
 - `deepseek` (OpenAI-compatible)

README.zh-CN.md

@@ -1,8 +1,14 @@
-# RexOS
+# LoopForge
 
 [English](README.md) | 简体中文
 
-RexOS 是一个长任务的 Agent OS：持久化记忆、工具沙盒、模型路由，以及一个 Anthropic 风格的 Harness，用于跨多次会话持续推进任务。
+LoopForge（原 RexOS）是一个长任务的 Agent OS：持久化记忆、工具沙盒、模型路由，以及一个 Anthropic 风格的 Harness，用于跨多次会话持续推进任务。
+
+## 品牌更新
+
+- 对外产品名：**LoopForge**
+- 兼容名仍保留：`rexos`（CLI）、`~/.rexos`（配置/数据目录）、`rexleimo/rexos`（仓库路径）
+- 现有脚本和文档中的 `rexos` 命令可继续使用
 
 ## 文档
 
@@ -39,13 +45,13 @@ cargo build --release -p rexos-cli
 
 ## 使用 Ollama（OpenAI 兼容）
 
-默认配置会在 `~/.rexos/config.toml` 里把 `ollama` 指向 `http://127.0.0.1:11434/v1`。
+LoopForge 默认配置会在 `~/.rexos/config.toml` 里把 `ollama` 指向 `http://127.0.0.1:11434/v1`。
 
 ```bash
 # 1) 启动 Ollama
 ollama serve
 
-# 2) 初始化 RexOS（创建 ~/.rexos/config.toml + ~/.rexos/rexos.db）
+# 2) 初始化 LoopForge（兼容命令：rexos）
 rexos init
 
 # 3) 在某个 workspace 目录里运行一次 agent session
@@ -69,7 +75,7 @@ git push origin v0.1.0
 
 ## Providers 与路由
 
-RexOS 通过多种 driver 支持多个 LLM Provider：
+LoopForge 通过多种 driver 支持多个 LLM Provider：
 - `openai_compatible`（Ollama / DeepSeek / Kimi / Qwen / GLM / MiniMax / NVIDIA NIM / 其它 OpenAI-compatible 网关）
 - `dashscope_native`（阿里云 DashScope Generation API / Qwen 原生）
 - `zhipu_native`（智谱 GLM 原生：auth/token 处理）
@@ -97,7 +103,7 @@ provider = "ollama"
 model = "default" # uses providers.<name>.default_model
 ```
 
-切换 provider：配置对应 provider 的 `api_key_env`（如需），并把 `[router.*]` 指向你想用的 provider；如果 `model = "default"`，RexOS 会使用 `providers.<name>.default_model`。
+切换 provider：配置对应 provider 的 `api_key_env`（如需），并把 `[router.*]` 指向你想用的 provider；如果 `model = "default"`，LoopForge 会使用 `providers.<name>.default_model`。
 
 内置 presets 包含：
 - `deepseek`（OpenAI-compatible）

docs-site/assets/images/favicon.svg

@@ -4,12 +4,14 @@ Practical notes for builders choosing an Agent OS and scaling from local prototy
 
 ## Latest posts
 
-- [RexOS vs OpenFang/OpenClaw (Builder View)](rexos-vs-openfang-openclaw.md)
+- [What Is LoopForge? (Formerly RexOS)](what-is-loopforge.md)
+- [LoopForge vs OpenFang/OpenClaw (Builder View)](rexos-vs-openfang-openclaw.md)
 - [Editorial Calendar](editorial-calendar.md)
 
 ## Suggested reading order
 
-1. Positioning and fit: [RexOS vs OpenFang/OpenClaw (Builder View)](rexos-vs-openfang-openclaw.md)
+1. Brand + product fit: [What Is LoopForge? (Formerly RexOS)](what-is-loopforge.md)
 2. First successful run: [New User Walkthrough](../tutorials/new-user-walkthrough.md)
 3. Practical templates: [10 Copy/Paste Tasks](../examples/case-tasks/ten-copy-paste-tasks.md)
-4. Publishing rhythm: [Editorial Calendar](editorial-calendar.md)
+4. Positioning and alternatives: [LoopForge vs OpenFang/OpenClaw (Builder View)](rexos-vs-openfang-openclaw.md)
+5. Publishing rhythm: [Editorial Calendar](editorial-calendar.md)

docs-site/assets/images/logo.svg

@@ -1,27 +1,27 @@
-# RexOS vs OpenFang/OpenClaw (Builder View)
+# LoopForge vs OpenFang/OpenClaw (Builder View)
 
 This comparison is based on local snapshots in this workspace on **March 4, 2026**.
 Goal: help builders choose the right tool faster.
 
 ## TL;DR
 
-- Choose **RexOS** if your core job is coding workflows with reproducible checkpoints and local-first model routing.
+- Choose **LoopForge** (formerly RexOS) if your core job is coding workflows with reproducible checkpoints and local-first model routing.
 - Choose **OpenFang** if your core job is multi-channel operations and autonomous "hands" at scale.
 - Choose **OpenClaw** if your core job is personal assistant experiences across many channels and devices.
 
 ## What each project optimizes for
 
 | Project | Strongest fit | Why |
 |---|---|---|
-| RexOS | Dev workflows + long-running engineering tasks | Harness loop (`change -> verify -> checkpoint`), workspace sandbox, SQLite memory, CLI-first |
+| LoopForge | Dev workflows + long-running engineering tasks | Harness loop (`change -> verify -> checkpoint`), workspace sandbox, SQLite memory, CLI-first |
 | OpenFang | Multi-agent operations + channel adapters | Heavy emphasis on channels, templates, operational breadth |
 | OpenClaw | Personal assistant platform + device/channel surfaces | Massive docs and channel coverage, onboarding wizard, broad ecosystem integration |
 
 ## Practical differences for builders
 
 ### 1) Reproducibility loop
 
-RexOS pushes a strict engineering loop:
+LoopForge pushes a strict engineering loop:
 
 ```bash
 rexos harness init my-repo
@@ -32,7 +32,7 @@ If your team measures progress by repeatable checkpoints and artifact trails, th
 
 ### 2) Local-first bring-up
 
-RexOS default path is optimized for local bring-up with Ollama:
+LoopForge default path is optimized for local bring-up with Ollama:
 
 ```bash
 ollama serve
@@ -48,7 +48,7 @@ What we learned from competitors:
 - OpenFang: clear install matrix + strong troubleshooting/FAQ structure.
 - OpenClaw: strong onboarding funnel + huge scenario inventory + broad links.
 
-What RexOS should do (and now starts doing):
+What LoopForge should do (and now starts doing):
 - add beginner FAQ
 - add more copy/paste task packs
 - add growth blog positioning pages
@@ -57,11 +57,11 @@ What RexOS should do (and now starts doing):
 
 Use this simple decision tree:
 
-1. Need coding workflow reliability first? -> **RexOS**
+1. Need coding workflow reliability first? -> **LoopForge**
 2. Need broad channel operations as first priority? -> **OpenFang**
 3. Need consumer-like personal assistant breadth? -> **OpenClaw**
 
-## Start with RexOS in 3 commands
+## Start with LoopForge in 3 commands
 
 ```bash
 ollama serve

docs-site/blog/index.md

@@ -0,0 +1,53 @@
+# What Is LoopForge? (Formerly RexOS)
+
+LoopForge is the new public brand of the product previously known as RexOS.
+We renamed to make the product easier to remember and easier to spread in builder communities.
+
+## One-line positioning
+
+LoopForge is a **local-first long-running Agent OS** for engineering workflows:
+`change -> verify -> checkpoint`.
+
+## Who it is for
+
+- Developers who want reproducible coding loops, not one-shot chat outputs.
+- Teams that need durable checkpoints, artifact trails, and workspace-safe tool execution.
+- Builders who start local with Ollama and later route to stronger cloud models.
+
+## What changed vs what stayed the same
+
+Changed:
+- Product/public name: **LoopForge**
+
+Unchanged for compatibility:
+- CLI command: `rexos`
+- Config and data dir: `~/.rexos`
+- Existing command snippets and scripts continue to work
+
+## Why the name "LoopForge"
+
+- `Loop`: long-running iterative workflows
+- `Forge`: build, shape, and harden software artifacts
+- Together: a toolchain identity that explains both process and outcome
+
+## 3-command first run
+
+```bash
+ollama serve
+rexos init
+rexos agent run --workspace loopforge-demo --prompt "Create notes/hello.md with a short intro to LoopForge."
+```
+
+## Brand keywords (for docs/search/community)
+
+- LoopForge
+- long-running agent OS
+- local-first coding agent
+- harness workflow
+- reproducible AI engineering loop
+
+## Next reading
+
+- [LoopForge vs OpenFang/OpenClaw (Builder View)](rexos-vs-openfang-openclaw.md)
+- [New User Walkthrough](../tutorials/new-user-walkthrough.md)
+- [10 Copy/Paste Tasks](../examples/case-tasks/ten-copy-paste-tasks.md)

docs-site/blog/rexos-vs-openfang-openclaw.md

@@ -1,8 +1,8 @@
 <div class="rexos-hero" markdown>
 
-# RexOS
+# LoopForge
 
-**Long-running Agent OS**: durable harness + SQLite memory + sandboxed tools + multi-provider routing.
+**Long-running Agent OS by Rex**: durable harness + SQLite memory + sandboxed tools + multi-provider routing.
 
 [Get started (Ollama)](tutorials/quickstart-ollama.md){ .md-button .md-button--primary }
 [Harness tutorial](tutorials/harness-long-task.md){ .md-button }
@@ -19,6 +19,8 @@ Develop locally with small models on Ollama, then switch routing to GLM / MiniMa
 
 </div>
 
+> Brand update: LoopForge is the public name (formerly RexOS). Compatibility remains unchanged for now: CLI is still `rexos`, and config/data path is still `~/.rexos`.
+
 <div class="grid cards" markdown>
 
 - :material-checklist: **Harness-first long tasks**  
@@ -48,7 +50,7 @@ Make sure you have at least one **chat model** available in Ollama (not embeddin
     # 1) Start Ollama
     ollama serve
 
-    # 2) Init RexOS (~/.rexos/config.toml + ~/.rexos/rexos.db)
+    # 2) Init LoopForge (compat command: rexos)
     rexos init
 
     # 3) Run a workspace session
@@ -61,7 +63,7 @@ Make sure you have at least one **chat model** available in Ollama (not embeddin
     # 1) Start Ollama
     ollama serve
 
-    # 2) Init RexOS (~/.rexos/config.toml + ~/.rexos/rexos.db)
+    # 2) Init LoopForge (compat command: rexos)
     rexos init
 
     # 3) Run a workspace session
@@ -73,7 +75,7 @@ Make sure you have at least one **chat model** available in Ollama (not embeddin
 
 ```mermaid
 flowchart LR
-  U[You] -->|prompt| R[RexOS CLI]
+  U[You] -->|prompt| R[LoopForge CLI]
   R --> M[(SQLite memory)]
   R --> W[(Workspace sandbox)]
   R --> G[(Git checkpoints)]
@@ -90,5 +92,6 @@ flowchart LR
 - Explore common recipes: `how-to/use-cases.md`
 - Start from beginner Q&A: `how-to/faq.md`
 - Copy/paste 10 practical tasks: `examples/case-tasks/ten-copy-paste-tasks.md`
+- Read the brand intro: `blog/what-is-loopforge.md`
 - Read the positioning article: `blog/rexos-vs-openfang-openclaw.md`
 - Switch providers (GLM/MiniMax native + NVIDIA NIM included): `how-to/providers.md`

docs-site/blog/what-is-loopforge.md

@@ -4,12 +4,14 @@
 
 ## 最新文章
 
-- [RexOS 与 OpenFang/OpenClaw 对比（开发者视角）](rexos-vs-openfang-openclaw.md)
+- [什么是 LoopForge？（原 RexOS）](what-is-loopforge.md)
+- [LoopForge 与 OpenFang/OpenClaw 对比（开发者视角）](rexos-vs-openfang-openclaw.md)
 - [内容发布日历](editorial-calendar.md)
 
 ## 建议阅读顺序
 
-1. 先看定位： [RexOS 与 OpenFang/OpenClaw 对比（开发者视角）](rexos-vs-openfang-openclaw.md)
+1. 先看品牌与定位： [什么是 LoopForge？（原 RexOS）](what-is-loopforge.md)
 2. 再跑通首条链路： [新人复习](../tutorials/new-user-walkthrough.md)
 3. 最后套模板实操： [10 个可复制任务](../examples/case-tasks/ten-copy-paste-tasks.md)
-4. 固定输出节奏： [内容发布日历](editorial-calendar.md)
+4. 对比竞品场景： [LoopForge 与 OpenFang/OpenClaw 对比（开发者视角）](rexos-vs-openfang-openclaw.md)
+5. 固定输出节奏： [内容发布日历](editorial-calendar.md)

docs-site/index.md

@@ -1,26 +1,26 @@
-# RexOS 与 OpenFang/OpenClaw 对比（开发者视角）
+# LoopForge 与 OpenFang/OpenClaw 对比（开发者视角）
 
 本文基于当前工作区在 **2026 年 3 月 4 日** 的本地快照做对比，目标是帮开发团队更快选型。
 
 ## 一句话结论
 
-- 如果你最关注“编码任务可复现 + 可追踪交付”，优先 **RexOS**。
+- 如果你最关注“编码任务可复现 + 可追踪交付”，优先 **LoopForge**（原 RexOS）。
 - 如果你最关注“多渠道运营 + 大量适配器”，优先 **OpenFang**。
 - 如果你最关注“个人助手体验 + 多端多渠道覆盖”，优先 **OpenClaw**。
 
 ## 三者各自最强项
 
 | 项目 | 最适合场景 | 核心原因 |
 |---|---|---|
-| RexOS | 工程研发流程、长任务推进 | Harness 循环（修改->验证->checkpoint）、workspace 沙盒、SQLite 记忆、CLI 优先 |
+| LoopForge | 工程研发流程、长任务推进 | Harness 循环（修改->验证->checkpoint）、workspace 沙盒、SQLite 记忆、CLI 优先 |
 | OpenFang | 多 Agent 运营和渠道分发 | 渠道覆盖与模板体系很重，偏运营执行面 |
 | OpenClaw | 个人助手产品形态 | 文档和渠道覆盖极广，上手向导完整 |
 
 ## 对研发团队最关键的差异
 
 ### 1）可复现交付链路
 
-RexOS 的核心优势是工程化循环：
+LoopForge 的核心优势是工程化循环：
 
 ```bash
 rexos harness init my-repo
@@ -31,7 +31,7 @@ rexos harness run my-repo --prompt "Run tests and fix one failing case"
 
 ### 2）本地起步成本
 
-RexOS 默认就是本地 Ollama 开发流：
+LoopForge 默认就是本地 Ollama 开发流：
 
 ```bash
 ollama serve
@@ -47,18 +47,18 @@ rexos agent run --workspace rexos-work --prompt "Create hello.txt with the word
 - OpenFang：安装矩阵清晰，FAQ/故障排查结构化。
 - OpenClaw：上手漏斗完整，场景文档数量巨大。
 
-RexOS 对应优化（本次已开始补齐）：
+LoopForge 对应优化（本次已开始补齐）：
 - 新手 FAQ
 - 更多可复制案例任务
 - 面向增长的对比博客
 
 ## 快速选型规则
 
-1. 首要目标是“研发稳定交付”？-> **RexOS**
+1. 首要目标是“研发稳定交付”？-> **LoopForge**
 2. 首要目标是“运营渠道铺开”？-> **OpenFang**
 3. 首要目标是“个人助手体验覆盖”？-> **OpenClaw**
 
-## 3 条命令先跑 RexOS
+## 3 条命令先跑 LoopForge
 
 ```bash
 ollama serve