refactor: rename project HexAgent → UniHarness

.gitattributes

@@ -1,5 +1,2 @@
-libs/openagent/sandbox/vm/setup/*.sh text eol=lf
-libs/openagent/sandbox/vm/setup/steps/*.sh text eol=lf
-libs/hexagent/sandbox/vm/setup/*.sh text eol=lf
-libs/hexagent/sandbox/vm/setup/steps/*.sh text eol=lf
-libs/openagent/sandbox/vm/wsl/prebuilt/openagent-prebuilt.tar filter=lfs diff=lfs merge=lfs -text
+libs/uniharness/sandbox/vm/setup/*.sh text eol=lf
+libs/uniharness/sandbox/vm/setup/steps/*.sh text eol=lf

.github/workflows/ci.yml

@@ -22,10 +22,10 @@ jobs:
       matrix:
         include:
           # Core package
-          - working-directory: "libs/hexagent"
+          - working-directory: "libs/uniharness"
             python-version: "3.11"
           # Other packages in monorepo
-          # - working-directory: "libs/hexagent-cli"
+          # - working-directory: "libs/uniharness-cli"
           #   python-version: "3.11"
       fail-fast: false
     uses: ./.github/workflows/_lint.yml
@@ -38,16 +38,16 @@ jobs:
       matrix:
         include:
           # Core package
-          - working-directory: "libs/hexagent"
+          - working-directory: "libs/uniharness"
             python-version: "3.11"
-          - working-directory: "libs/hexagent"
+          - working-directory: "libs/uniharness"
             python-version: "3.12"
-          - working-directory: "libs/hexagent"
+          - working-directory: "libs/uniharness"
             python-version: "3.13"
           # Other packages
-          # - working-directory: "libs/hexagent-cli"
+          # - working-directory: "libs/uniharness-cli"
           #   python-version: "3.11"
-          # - working-directory: "libs/hexagent-cli"
+          # - working-directory: "libs/uniharness-cli"
           #   python-version: "3.13"
       fail-fast: false
     uses: ./.github/workflows/_test.yml
@@ -59,7 +59,7 @@ jobs:
     strategy:
       matrix:
         include:
-          - working-directory: "libs/hexagent"
+          - working-directory: "libs/uniharness"
             python-version: "3.11"
       fail-fast: false
     uses: ./.github/workflows/_integration_test.yml

.gitignore

@@ -215,14 +215,14 @@ __marimo__/
 /tmp/
 */tmp/
 
-# HexAgent Sandbox
+# UniHarness Sandbox
 mounts/
 
 # Git worktrees
 .trees/
 
 # Electron build artifact
-libs/hexagent_demo/electron/dist.zip
+libs/uniharness_demo/electron/dist.zip
 
 # Vite dev-server cache
 **/.vite/

.pre-commit-config.yaml

@@ -17,9 +17,9 @@ repos:
 
   - repo: local
     hooks:
-      - id: hexagent
-        name: format and lint hexagent
+      - id: uniharness
+        name: format and lint uniharness
         language: system
-        entry: make -C libs/hexagent format lint
-        files: ^libs/hexagent/
+        entry: make -C libs/uniharness format lint
+        files: ^libs/uniharness/
         pass_filenames: false

CONTRIBUTING.md

@@ -1,6 +1,6 @@
-# Contributing to HexAgent
+# Contributing to UniHarness
 
-Thank you for your interest in contributing to HexAgent! We believe agent infrastructure should be open, vendor-neutral, and community-driven — and every contribution moves that forward.
+Thank you for your interest in contributing to UniHarness! We believe agent infrastructure should be open, vendor-neutral, and community-driven — and every contribution moves that forward.
 
 Whether you're fixing a typo, adding a new tool, implementing a computer backend, or improving documentation, you're helping build the agent harness the community needs.
 
@@ -16,11 +16,11 @@ Whether you're fixing a typo, adding a new tool, implementing a computer backend
 
 ```bash
 # Clone the repository
-git clone https://github.com/an7tang/hexagent.git
-cd hexagent
+git clone https://github.com/UnicomAI/UniHarness.git
+cd uniharness
 
 # Set up the core library
-cd libs/hexagent
+cd libs/uniharness
 uv sync --group test
 
 # Verify everything works
@@ -31,15 +31,15 @@ make test
 ### Project Structure
 
 ```
-hexagent/
+uniharness/
 ├── libs/
-│   ├── hexagent/          # Core agent harness library
-│   │   ├── hexagent/      #   Package source
+│   ├── uniharness/          # Core agent harness library
+│   │   ├── uniharness/      #   Package source
 │   │   ├── tests/         #   Unit + integration tests
 │   │   ├── sandbox/       #   Docker/VM sandbox configs
 │   │   ├── Makefile       #   Build targets
 │   │   └── pyproject.toml #   Package config
-│   └── hexagent_demo/     # Demo desktop application
+│   └── uniharness_demo/     # Demo desktop application
 │       ├── backend/       #   FastAPI backend
 │       ├── frontend/      #   React frontend
 │       └── electron/      #   Electron shell
@@ -52,7 +52,7 @@ hexagent/
 ### Running Tests
 
 ```bash
-cd libs/hexagent
+cd libs/uniharness
 
 make test              # Unit tests with coverage
 make integration_test  # Integration tests (requires API keys)
@@ -94,14 +94,14 @@ pre-commit install
 - Test behavior, not implementation details
 - Prefer testing public APIs over internal functions
 - Descriptive test names: `test_<action>_<condition>_<expected_result>`
-- Unit tests in `tests/unit_tests/` — mirror the `hexagent/` directory structure
+- Unit tests in `tests/unit_tests/` — mirror the `uniharness/` directory structure
 - Integration tests in `tests/integration_tests/`
 
 ## Making Changes
 
 ### Before You Start
 
-1. Check existing [issues](https://github.com/an7tang/hexagent/issues) to see if someone is already working on it
+1. Check existing [issues](https://github.com/UnicomAI/UniHarness/issues) to see if someone is already working on it
 2. For significant changes, open an issue first to discuss the approach
 3. Fork the repository and create a feature branch
 
@@ -116,7 +116,7 @@ pre-commit install
 
 3. Ensure all checks pass:
    ```bash
-   cd libs/hexagent
+   cd libs/uniharness
    make format
    make lint
    make test
@@ -151,31 +151,31 @@ When contributing, keep these principles in mind:
 
 ### Framework-Agnostic Core
 
-The core library (`hexagent/`) is framework-agnostic — LangChain integration lives in `hexagent/langchain/`. Don't introduce LangChain imports outside of that directory.
+The core library (`uniharness/`) is framework-agnostic — LangChain integration lives in `uniharness/langchain/`. Don't introduce LangChain imports outside of that directory.
 
 ### Where to Contribute
 
 | Area | Good for | Location |
 |------|----------|----------|
-| New tools | Adding capabilities to agents | `hexagent/tools/` |
-| Computer implementations | New execution environments | `hexagent/computer/` |
-| Web providers | New search/fetch backends | `hexagent/tools/web/providers/` |
-| MCP improvements | Protocol support | `hexagent/mcp/` |
-| Prompt fragments | Better agent instructions | `hexagent/prompts/fragments/` |
-| Demo features | UI/UX improvements | `libs/hexagent_demo/` |
+| New tools | Adding capabilities to agents | `uniharness/tools/` |
+| Computer implementations | New execution environments | `uniharness/computer/` |
+| Web providers | New search/fetch backends | `uniharness/tools/web/providers/` |
+| MCP improvements | Protocol support | `uniharness/mcp/` |
+| Prompt fragments | Better agent instructions | `uniharness/prompts/fragments/` |
+| Demo features | UI/UX improvements | `libs/uniharness_demo/` |
 | Tests | Improving coverage | `tests/` |
 | Documentation | Clarity and examples | `README.md`, `libs/*/README.md` |
 
 ## Good First Issues
 
-New to the project? Look for issues tagged [`good first issue`](https://github.com/an7tang/hexagent/labels/good%20first%20issue). These are scoped to be approachable without deep knowledge of the codebase.
+New to the project? Look for issues tagged [`good first issue`](https://github.com/UnicomAI/UniHarness/labels/good%20first%20issue). These are scoped to be approachable without deep knowledge of the codebase.
 
 ## Reporting Issues
 
-- Use [GitHub Issues](https://github.com/an7tang/hexagent/issues)
+- Use [GitHub Issues](https://github.com/UnicomAI/UniHarness/issues)
 - Include steps to reproduce, expected behavior, and actual behavior
 - For bugs, include your Python version, OS, and relevant dependency versions
 
 ## Questions?
 
-Open a [Discussion](https://github.com/an7tang/hexagent/discussions) for questions, ideas, or feedback that don't fit neatly into an issue.
+Open a [Discussion](https://github.com/UnicomAI/UniHarness/discussions) for questions, ideas, or feedback that don't fit neatly into an issue.

README.md

@@ -1,41 +1,41 @@
 <p align="center">
-  <img src=".github/images/logo.svg" alt="HexAgent" width="300"/>
+  <img src=".github/images/logo.svg" alt="UniHarness" width="300"/>
 </p>
 
 <p align="center">
   <strong>Agent harness — give any LLM a computer. Ship any agent product.</strong>
 </p>
 
 <p align="center">
-  <a href="https://github.com/an7tang/hexagent/actions/workflows/ci.yml"><img src="https://github.com/an7tang/hexagent/actions/workflows/ci.yml/badge.svg" alt="CI"></a>
+  <a href="https://github.com/UnicomAI/UniHarness/actions/workflows/ci.yml"><img src="https://github.com/UnicomAI/UniHarness/actions/workflows/ci.yml/badge.svg" alt="CI"></a>
   <a href="https://www.python.org/downloads/"><img src="https://img.shields.io/badge/python-3.11%2B-blue" alt="Python 3.11+"></a>
   <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-green" alt="License: MIT"></a>
 </p>
 
 ---
 
-**HexAgent** is an open-source **agent harness**: the production runtime that gives any LLM a fully-equipped computer — terminal, filesystem, and shell — to complete tasks autonomously.
+**UniHarness** is an open-source **agent harness**: the production runtime that gives any LLM a fully-equipped computer — terminal, filesystem, and shell — to complete tasks autonomously.
 
-Unlike every other agent framework, HexAgent **separates the agent runtime from the computer it operates on**. Your agent gets a sandboxed machine; your runtime keeps its API keys, config, and source code private.
+Unlike every other agent framework, UniHarness **separates the agent runtime from the computer it operates on**. Your agent gets a sandboxed machine; your runtime keeps its API keys, config, and source code private.
 
 > **Why "harness" and not "framework"?** A framework gives you building blocks and says "assemble your own agent." A harness gives the agent a fully equipped runtime — tools, context management, safety, execution environments — so you focus on *what* the agent does, not *how* it executes. ([Read more](#agent-harness-vs-agent-framework))
 
 ## The Computer Layer
 
-In Claude Code, Codex, and every LangChain agent, the agent runtime and the computer it controls are the same process. The agent can read its own source code, config files, and API keys. HexAgent's `Computer` protocol makes this separation explicit and pluggable — swap execution environments without changing a line of agent code.
+In Claude Code, Codex, and every LangChain agent, the agent runtime and the computer it controls are the same process. The agent can read its own source code, config files, and API keys. UniHarness's `Computer` protocol makes this separation explicit and pluggable — swap execution environments without changing a line of agent code.
 
 ```python
-from hexagent import create_agent
-from hexagent.computer import LocalNativeComputer, LocalVM, RemoteE2BComputer
+from uniharness import create_agent
+from uniharness.computer import LocalNativeComputer, LocalVM, RemoteE2BComputer
 
 # Development — run on your machine
-agent = await create_agent(model="anthropic/claude-sonnet-4-20250514", computer=LocalNativeComputer())
+agent = await create_agent(model="openai:gpt-5.5", computer=LocalNativeComputer())
 
 # Security-sensitive — sandboxed VM (Lima on macOS, WSL on Windows)
-agent = await create_agent(model="anthropic/claude-sonnet-4-20250514", computer=LocalVM())
+agent = await create_agent(model="openai:gpt-5.5", computer=LocalVM())
 
 # Production / multi-tenant — isolated cloud sandbox
-agent = await create_agent(model="anthropic/claude-sonnet-4-20250514", computer=RemoteE2BComputer(api_key="..."))
+agent = await create_agent(model="openai:gpt-5.5", computer=RemoteE2BComputer(api_key="..."))
 ```
 
 ```
@@ -64,19 +64,19 @@ Implement the `Computer` protocol to add your own — Docker, Kubernetes pods, o
 ## Quick Start
 
 ```bash
-pip install hexagent
+pip install uniharness
 ```
 
 ### Minimal example
 
 ```python
 import asyncio
-from hexagent import create_agent
-from hexagent.computer import LocalNativeComputer
+from uniharness import create_agent
+from uniharness.computer import LocalNativeComputer
 
 async def main():
     async with await create_agent(
-        model="anthropic/claude-sonnet-4-20250514",  # or any LLM
+        model="openai:gpt-5.5",  # or any LLM
         computer=LocalNativeComputer(),
     ) as agent:
         result = await agent.ainvoke({
@@ -90,12 +90,12 @@ asyncio.run(main())
 ### Use any model
 
 ```python
-from hexagent import create_agent, ModelProfile
-from hexagent.computer import LocalNativeComputer
+from uniharness import create_agent, ModelProfile
+from uniharness.computer import LocalNativeComputer
 
 # DeepSeek, Qwen, Llama, Mistral — anything OpenAI-compatible
 model = ModelProfile(
-    model="deepseek/deepseek-chat",
+    model="deepseek:deepseek-v4-flash",
     base_url="https://api.deepseek.com/v1",
     api_key="your-key",
     context_window=64000,
@@ -107,11 +107,11 @@ agent = await create_agent(model=model, computer=LocalNativeComputer())
 ### Add subagents, MCP servers, web tools
 
 ```python
-from hexagent import create_agent, AgentDefinition
-from hexagent.computer import LocalNativeComputer
+from uniharness import create_agent, AgentDefinition
+from uniharness.computer import LocalNativeComputer
 
 agent = await create_agent(
-    model="anthropic/claude-sonnet-4-20250514",
+    model="openai:gpt-5.5",
     computer=LocalNativeComputer(),
     # Subagents for parallel specialized work
     agents={
@@ -134,17 +134,17 @@ agent = await create_agent(
 ### Run in a cloud sandbox
 
 ```python
-from hexagent import create_agent
-from hexagent.computer import RemoteE2BComputer
+from uniharness import create_agent
+from uniharness.computer import RemoteE2BComputer
 
 # Fully isolated cloud execution — no local risk
 agent = await create_agent(
-    model="openai/gpt-4o",
+    model="openai:gpt-4o",
     computer=RemoteE2BComputer(api_key="your-e2b-key"),
 )
 ```
 
-See [`libs/hexagent/README.md`](libs/hexagent/README.md) for the full API reference.
+See [`libs/uniharness/README.md`](libs/uniharness/README.md) for the full API reference.
 
 ## What Can You Build?
 
@@ -157,7 +157,7 @@ One harness powers four product types — no other agent SDK does this:
 | **Cowork** | Desktop agent that works on your local files, folders, and apps — completing knowledge work tasks autonomously while you steer | [Claude Cowork](https://www.anthropic.com/product/claude-cowork) |
 | **Autonomous Agent** | Headless agent that runs tasks end-to-end without supervision | [OpenClaw](https://github.com/openclaw/openclaw), [Devin](https://devin.ai/) |
 
-The [`hexagent_demo`](libs/hexagent_demo/) app ships with ready-to-use **Chat** and **Cowork** modes as concrete examples.
+The [`uniharness_demo`](libs/uniharness_demo/) app ships with ready-to-use **Chat** and **Cowork** modes as concrete examples.
 
 ## Features
 
@@ -178,9 +178,9 @@ The [`hexagent_demo`](libs/hexagent_demo/) app ships with ready-to-use **Chat**
 - **Web providers** — Pluggable search (Tavily/Brave) and fetch (Jina/Firecrawl) backends
 - **Composable, not magical** — Small modules with explicit I/O. No hidden state. Every piece is testable and replaceable.
 
-### How HexAgent Compares
+### How UniHarness Compares
 
-| | HexAgent | Claude Agent SDK | LangChain Deep Agents |
+| | UniHarness | Claude Agent SDK | LangChain Deep Agents |
 |---|---|---|---|
 | **Open source** | MIT | MIT | MIT |
 | **Model-agnostic** | Any LLM | Claude only | Any LLM |
@@ -243,23 +243,23 @@ The AI agent ecosystem has converged on a [clear taxonomy](https://blog.langchai
 | **What it is** | Building blocks (tools, prompts, memory) | Durable execution engine | Complete agent operating system |
 | **Analogy** | A toolkit | A job scheduler | An OS for the agent |
 | **You build** | Everything from scratch | Orchestration logic | Your agent's purpose |
-| **Examples** | LangChain, CrewAI, Semantic Kernel | LangGraph, Temporal | HexAgent, Claude Code, OpenHands |
+| **Examples** | LangChain, CrewAI, Semantic Kernel | LangGraph, Temporal | UniHarness, Claude Code, OpenHands |
 
 A framework says: *"Here are components. Assemble your agent."*
 
 A harness says: *"Here is a fully equipped computer. Tell the agent what to do."*
 
-HexAgent is a harness you can embed as a library — giving you the batteries-included runtime of products like Claude Code, with the flexibility to build any agent product you want.
+UniHarness is a harness you can embed as a library — giving you the batteries-included runtime of products like Claude Code, with the flexibility to build any agent product you want.
 
 ## Project Structure
 
 | Package | Description |
 |---------|-------------|
-| [`libs/hexagent`](libs/hexagent/) | **Core framework** — the agent harness library ([API docs](libs/hexagent/README.md)) |
-| [`libs/hexagent_demo`](libs/hexagent_demo/) | **Demo app** — desktop Chat + Cowork built on the framework ([setup guide](libs/hexagent_demo/README.md)) |
+| [`libs/uniharness`](libs/uniharness/) | **Core framework** — the agent harness library ([API docs](libs/uniharness/README.md)) |
+| [`libs/uniharness_demo`](libs/uniharness_demo/) | **Demo app** — desktop Chat + Cowork built on the framework ([setup guide](libs/uniharness_demo/README.md)) |
 
 ```
-libs/hexagent/hexagent/
+libs/uniharness/uniharness/
 ├── computer/       # Computer protocol — local, VM, or cloud (E2B) execution
 ├── harness/        # Runtime augmentation: environment, permissions, skills, reminders
 ├── tools/          # Built-in tools: CLI, web, subagents, skills, todos