feat: new version more robust network handle

* feat: added e2e

* fix: try fix

* fix: try fix

* fix: try fix

* fix: added more observability

* ci: fixed double trigger

* feat: partial e2e test with mock vm

* feat: more monkey tests

* feat: added coverage

.github/workflows/ci.yml

@@ -10,7 +10,6 @@ on:
       - "README.zh.md"
       - "docs/**"
       - "install"
-  pull_request:
 
 concurrency:
   group: ci-${{ github.ref }}
@@ -20,6 +19,8 @@ jobs:
   fmt:
     name: Format
     runs-on: macos-latest
+    env:
+      RUST_BACKTRACE: "1"
     steps:
       - uses: actions/checkout@v6
       - name: Install Rust
@@ -34,6 +35,8 @@ jobs:
   clippy:
     name: Clippy
     runs-on: macos-latest
+    env:
+      RUST_BACKTRACE: "1"
     steps:
       - uses: actions/checkout@v6
       - name: Install Rust
@@ -48,6 +51,8 @@ jobs:
   build:
     name: Build
     runs-on: macos-latest
+    env:
+      RUST_BACKTRACE: "1"
     steps:
       - uses: actions/checkout@v6
       - name: Install Rust
@@ -60,6 +65,8 @@ jobs:
   test:
     name: Test
     runs-on: macos-latest
+    env:
+      RUST_BACKTRACE: "full"
     steps:
       - uses: actions/checkout@v6
       - name: Install Rust
@@ -68,3 +75,24 @@ jobs:
         uses: Swatinem/rust-cache@v2
       - name: cargo test
         run: cargo test --locked
+
+  coverage:
+    name: Coverage
+    runs-on: macos-latest
+    needs: [fmt, clippy, build, test]
+    env:
+      RUST_BACKTRACE: "full"
+    steps:
+      - uses: actions/checkout@v6
+      - name: Install Rust
+        uses: dtolnay/rust-toolchain@stable
+        with:
+          components: llvm-tools-preview
+      - name: Cache Rust
+        uses: Swatinem/rust-cache@v2
+      - name: Install cargo-llvm-cov
+        uses: taiki-e/install-action@v2
+        with:
+          tool: cargo-llvm-cov
+      - name: cargo llvm-cov
+        run: cargo llvm-cov --locked --features mock-vm --tests -- --nocapture

Cargo.lock

@@ -82,6 +82,21 @@ dependencies = [
  "windows-sys 0.61.2",
 ]
 
+[[package]]
+name = "assert_cmd"
+version = "2.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9c5bcfa8749ac45dd12cb11055aeeb6b27a3895560d60d71e3c23bf979e60514"
+dependencies = [
+ "anstyle",
+ "bstr",
+ "libc",
+ "predicates",
+ "predicates-core",
+ "predicates-tree",
+ "wait-timeout",
+]
+
 [[package]]
 name = "backtrace"
 version = "0.3.76"
@@ -112,6 +127,17 @@ dependencies = [
  "objc2",
 ]
 
+[[package]]
+name = "bstr"
+version = "1.12.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "63044e1ae8e69f3b5a92c736ca6269b8d12fa7efe39bf34ddb06d102cf0e2cab"
+dependencies = [
+ "memchr",
+ "regex-automata",
+ "serde",
+]
+
 [[package]]
 name = "bumpalo"
 version = "3.19.1"
@@ -321,6 +347,12 @@ dependencies = [
  "zeroize",
 ]
 
+[[package]]
+name = "difflib"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6184e33543162437515c2e2b48714794e37845ec9851711914eec9d308f6ebe8"
+
 [[package]]
 name = "dispatch2"
 version = "0.3.0"
@@ -737,6 +769,33 @@ version = "0.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "439ee305def115ba05938db6eb1644ff94165c5ab5e9420d1c1bcedbba909391"
 
+[[package]]
+name = "predicates"
+version = "3.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a5d19ee57562043d37e82899fade9a22ebab7be9cef5026b07fda9cdd4293573"
+dependencies = [
+ "anstyle",
+ "difflib",
+ "predicates-core",
+]
+
+[[package]]
+name = "predicates-core"
+version = "1.0.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "727e462b119fe9c93fd0eb1429a5f7647394014cf3c04ab2c0350eeb09095ffa"
+
+[[package]]
+name = "predicates-tree"
+version = "1.0.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "72dd2d6d381dfb73a193c7fca536518d7caee39fc8503f74e7dc0be0531b425c"
+dependencies = [
+ "predicates-core",
+ "termtree",
+]
+
 [[package]]
 name = "proc-macro2"
 version = "1.0.106"
@@ -1007,6 +1066,12 @@ dependencies = [
  "windows-sys 0.61.2",
 ]
 
+[[package]]
+name = "termtree"
+version = "0.5.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8f50febec83f5ee1df3015341d8bd429f2d1cc62bcba7ea2076759d315084683"
+
 [[package]]
 name = "thiserror"
 version = "2.0.18"
@@ -1238,8 +1303,9 @@ checksum = "ba73ea9cf16a25df0c8caa16c51acb937d5712a8429db78a3ee29d5dcacd3a65"
 
 [[package]]
 name = "vibebox"
-version = "0.2.2"
+version = "0.2.5"
 dependencies = [
+ "assert_cmd",
  "block2",
  "clap",
  "color-eyre",
@@ -1261,6 +1327,15 @@ dependencies = [
  "uuid",
 ]
 
+[[package]]
+name = "wait-timeout"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "09ac3b126d3914f9849036f826e054cbabdc8519970b8998ddaf3b5bd3c65f11"
+dependencies = [
+ "libc",
+]
+
 [[package]]
 name = "wasi"
 version = "0.11.1+wasi-snapshot-preview1"

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "vibebox"
-version = "0.2.2"
+version = "0.2.5"
 edition = "2024"
 authors = ["Finn Sheng"]
 description = "Ultrafast CLI on Apple Silicon macOS for fast, sandboxed development and LLM agents."
@@ -42,3 +42,10 @@ ratatui = { version = "0.29.0", features = ["unstable-rendered-line-info"] }
 tracing = "0.1"
 tracing-subscriber = { version = "0.3", features = ["env-filter"] }
 dialoguer = "0.12.0"
+
+[dev-dependencies]
+assert_cmd = "2"
+tempfile = "3"
+
+[features]
+mock-vm = []

README.md

@@ -5,7 +5,7 @@
     </picture>
   </a>
 </p>
-<p align="center">Your ultrafast open source AI sandbox.</p>
+<p align="center">an ultrafast, open-source sandbox for running coding agents safely.</p>
 
 <p align="center">
   <a href="https://crates.io/crates/vibebox">
@@ -24,33 +24,101 @@
   <a href="README.zh.md">简体中文</a>
 </p>
 
-[![OpenCode Terminal UI](docs/screenshot.png)](https://vibebox.robcholz.com)
+**VibeBox is a per-project micro-VM sandbox for running coding agents on macOS (Apple Virtualization Framework).**
+It’s optimized for a *daily-driver* workflow: fast warm re-entry, explicit mounts, and reusable sessions.
+
+**Who it’s for:** macOS users running coding agents who want real isolation without giving up a fast daily workflow.
+
+**Quick facts:** warm re-entry is typically **<5s** on my M3 (varies by machine/cache); first run downloads and
+provisions a Debian base image (network dependent).
+
+**Security model:** Linux guest VM with explicit mount allowlists from `vibebox.toml` (repo-first, everything else
+opt-in).
+
+- **enter/attach in seconds:** `vibebox` drops you into a reusable sandbox for the current repo
+- **project-scoped by default:** explicit mounts + repo-contained changes (repo-first, everything else is allowlisted)
+- **sessioned:** multi-instance + session management (reuse, multiple terminals, cleanup)
+
+### Quick Demo
+
+```bash
+# from any repo
+cd my-project
+vibebox
+```
+
+What you should see (roughly):
+
+```text
+vibebox: starting (session: my-project)
+vibebox: attaching...
+vibecoder@vibebox:~/my-project$
+```
+
+[![VibeBox Terminal UI](docs/screenshot.png)](https://vibebox.robcholz.com)
 
 ---
 
+### Why I built VibeBox
+
+I use coding agents daily, and I wanted to give them a real shell without handing them my host machine.
+Lock things down and you get nonstop confirmations; loosen it up and you worry about deleting files, touching secrets,
+or wandering outside the repo.
+
+VibeBox is the middle ground: a per-repo sandbox with a hard VM boundary, fast re-entry, and explicit mounts. It’s built
+to be “always on” for agent work without turning safety into a chore.
+
+### Why a micro-VM (vs containers)?
+
+Containers are great. VibeBox isn’t trying to replace Docker/devcontainers for building services.
+
+I specifically wanted a VM-shaped default for agent workflows on macOS:
+
+- **guest-kernel isolation boundary by default:** when I’m letting an agent run arbitrary commands, I want “safe mode”
+  to
+  be a Linux guest, not my host.
+- **sessions as a first-class workflow:** attach/reuse per repo, multiple terminals into the same sandbox, reliable
+  cleanup to avoid orphan environments.
+- **explicit mount allowlists as the primary UX:** repo-scoped by default; anything else is an explicit decision.
+- **minimal per-repo setup:** you *can* reproduce parts of this with compose/devcontainers, but I wanted a single
+  command
+  that works repo-to-repo without maintaining container configs for the basic “safe shell” workflow.
+
+### Comparison
+
+Here’s why I didn’t just use existing options:
+
+- **vibe**: super convenient and nails “zero-config, just go”. VibeBox is intentionally on a different axis: per-repo
+  config + sessions + multi-instance lifecycle.
+- **QEMU**: powerful, but the configuration surface area is huge. For day-to-day sandboxing it’s not “open a repo and
+  go” — it’s a project on its own.
+- **Docker / devcontainers / devpods**: great ecosystem. My friction wasn’t raw startup time, it was the day-to-day
+  overhead of keeping per-repo agent sandboxes *safe-by-default* (mount allowlists, secrets exposure, attach/reuse,
+  cleanup) without maintaining container configs per repo for the basic workflow.
+
+That’s what pushed me to build **VibeBox**: I wanted a per-project sandbox that’s fast to enter (just `vibebox`),
+supports real configuration + sessions, and keeps a hard isolation boundary.
+
 ### Installation
 
 ```bash
-# YOLO
+# install script
 curl -fsSL https://raw.githubusercontent.com/robcholz/vibebox/main/install | bash
 
-# Package managers
+# package managers
 cargo install vibebox
 
-# Or manually (bad)
+# manual install
 curl -LO https://github.com/robcholz/vibebox/releases/download/latest/vibebox-macos-arm64.zip
 unzip vibebox-macos-arm64.zip
 mkdir -p ~/.local/bin
-mv vibe ~/.local/bin
+mv vibebox ~/.local/bin
 export PATH="$HOME/.local/bin:$PATH"
 ```
 
-> [!TIP]
-> We truly recommend you to use `YOLO` to install.
-
 **Requirements**
 
-- macOS on Apple Silicon (Vibebox uses Apple's virtualization APIs).
+- macOS on Apple Silicon (VibeBox uses Apple's virtualization APIs).
 
 **First Run**
 
@@ -66,7 +134,7 @@ cd /path/to/your/project
 vibebox
 ```
 
-On first run, Vibebox creates `vibebox.toml` in your project (if missing) and a `.vibebox/` directory for instance data.
+On first run, VibeBox creates `vibebox.toml` in your project (if missing) and a `.vibebox/` directory for instance data.
 
 **Configuration (`vibebox.toml`)**
 
@@ -117,7 +185,7 @@ vibebox explain     # show mounts and network info
 - Default SSH user: `vibecoder`
 - Hostname: `vibebox`
 - Base image provisioning installs: build tools, `git`, `curl`, `ripgrep`, `openssh-server`, and `sudo`.
-- On first login, Vibebox installs `mise` and configures tools like `uv`, `node`, `@openai/codex`, and
+- On first login, VibeBox installs `mise` and configures tools like `uv`, `node`, `@openai/codex`, and
   `@anthropic-ai/claude-code` (best-effort).
 - Shell aliases: `:help` and `:exit`.
 
@@ -132,27 +200,24 @@ vibebox explain     # show mounts and network info
 If you're interested in contributing to VibeBox, please read our [contributing docs](CONTRIBUTING.md) before
 submitting a pull request.
 
-### Using VibeBox
-
-Feel free to use, but remember to promote VibeBox as well!
-
 ### FAQ
 
-#### How is this different from other Sandboxes?
+#### How is this different from other sandboxes?
 
-Vibebox is built for fast, repeatable local sandboxes with minimal ceremony. What’s different here:
+VibeBox is built for fast, repeatable local sandboxes with minimal ceremony. What’s different here:
 
-- Warm startup is typically under **6 seconds** on my M3, so you can jump back in quickly.
+- Warm re-entry is typically **<5s** on my M3 (varies by machine/cache), so you can jump back in quickly.
 - One simple command — `vibebox` — drops you into the sandbox from your project.
 - Configuration lives in `vibebox.toml`, where you can set CPU, RAM, disk size, and mounts.
+- Sessions are first-class: reuse, multiple terminals, cleanup.
 
-### Special Thank
+### Special thanks
 
 [vibe](https://github.com/lynaghk/vibe) by lynaghk.
 
-And amazing Rust community, without your rich crates and fantastic toolchain like [crates.io](https://crates.io), this
-wouldn't be possible!
+And the amazing Rust community — without the ecosystem and toolchain like [crates.io](https://crates.io), this wouldn't
+be possible!
 
 ---
 
-**Follow me on X** [X.com](https://x.com/robcholz)
+**Follow me on X** [x.com/robcholz](https://x.com/robcholz)

README.zh.md

@@ -5,7 +5,7 @@
     </picture>
   </a>
 </p>
-<p align="center">超高速、开源的 AI 沙盒。</p>
+<p align="center">用于安全运行 coding agents 的超高速开源沙盒。</p>
 
 <p align="center">
   <a href="https://crates.io/crates/vibebox">
@@ -24,30 +24,86 @@
   <a href="README.md">English</a>
 </p>
 
-[![OpenCode Terminal UI](docs/screenshot.png)](https://vibebox.robcholz.com)
+**VibeBox 是一个按项目划分的 micro-VM 沙盒，用于在 macOS 上运行 coding agents（基于 Apple Virtualization Framework）。**
+它面向 *日常使用* 工作流优化：快速热启动、显式挂载、可复用会话。
+
+**适合谁：** 在 macOS 上使用 coding agents，并且既想要真实隔离又不想牺牲日常效率的人。
+
+**快速事实：** 在我的 M3 上，热启动通常 **<5s**（因机器/缓存而异）；首次运行会下载并初始化 Debian 基础镜像（受网络影响）。
+
+**安全模型：** Linux 来宾 VM + `vibebox.toml` 显式挂载白名单（默认仅项目目录，其它均需显式允许）。
+
+- **几秒进入/附加：** `vibebox` 直接进入当前仓库的可复用沙盒
+- **默认按项目范围：** 显式挂载 + 改动限制在仓库内（repo 优先，其他需 allowlist）
+- **会话化：** 多实例 + 会话管理（复用、多终端、清理）
+
+### 快速演示
+
+```bash
+# 在任意仓库内
+cd my-project
+vibebox
+```
+
+你大致会看到：
+
+```text
+vibebox: starting (session: my-project)
+vibebox: attaching...
+vibecoder@vibebox:~/my-project$
+```
+
+[![VibeBox Terminal UI](docs/screenshot.png)](https://vibebox.robcholz.com)
 
 ---
 
+### 我为什么做 VibeBox
+
+我每天都在用 coding agents，也希望它们有一个真实的 shell，但不想把宿主机直接交出去。
+权限收紧会被不停的确认打断；权限放开又担心误删文件、触及密钥，或者跑出仓库边界。
+
+VibeBox 是中间方案：按项目隔离、硬 VM 边界、快速回到工作状态、显式挂载。它适合把 agent 当成日常工具，而不是把安全变成负担。
+
+### 为什么是 micro-VM（而不是容器）？
+
+容器很好用。VibeBox 并不是用来替代 Docker/devcontainers 去构建服务。
+
+我更想要的是在 macOS 上适合 agent 的 VM 默认形态：
+
+- **默认是 guest-kernel 隔离边界：** 让 agent 跑任意命令时，“安全模式”是 Linux 来宾而不是宿主机。
+- **会话是第一等公民：** 按项目附加/复用，多终端进入同一沙盒，可靠清理，避免孤儿环境。
+- **显式挂载白名单作为主要 UX：** 默认仅项目目录，其它都需显式允许。
+- **最少的每项目配置：** 你可以用 compose/devcontainers 实现一部分，但我想要的是一个命令，在不同仓库间直接工作，不必维护容器配置来获得基础“安全
+  shell”体验。
+
+### 对比
+
+下面是我为什么没有直接用现成方案的原因：
+
+- **vibe**：非常方便，“零配置、直接用”做得很好。但 VibeBox 走的是另一条路：按项目配置 + 会话 + 多实例生命周期。
+- **QEMU**：很强大，但配置面太大了。日常当沙箱用，它不像是“进到 repo 就能用”，更像是你得先把它当成一个项目来折腾。
+- **Docker / devcontainers / devpods**：生态很成熟。我的痛点不在启动时间，而是日常保持
+  safe-by-default（挂载白名单、密钥暴露、附加/复用、清理）的开销，不想为基础 workflow 在每个项目维护容器配置。
+
+这就是我做 **VibeBox** 的原因：我想要一个按项目隔离的沙箱，进入快（直接 `vibebox`），支持真实配置 + 会话，同时保持硬隔离边界。
+
 ### 安装
 
 ```bash
-# YOLO：一键安装（推荐）
+# YOLO：一键安装
 curl -fsSL https://raw.githubusercontent.com/robcholz/vibebox/main/install | bash
 
 # Cargo
 cargo install vibebox
 
-# 或者手动安装（不推荐）
+# 或者手动安装
 curl -LO https://github.com/robcholz/vibebox/releases/download/latest/vibebox-macos-arm64.zip
 unzip vibebox-macos-arm64.zip
 mkdir -p ~/.local/bin
-mv vibe ~/.local/bin
+mv vibebox ~/.local/bin
 export PATH="$HOME/.local/bin:$PATH"
 ```
 
-> [!TIP]
-> 强烈建议直接用 `YOLO` 方式安装，省事且更不容易踩坑。
-
 **系统要求**
 
 - Apple Silicon 的 macOS（VibeBox 使用了 Apple 的虚拟化 API）。
@@ -130,19 +186,16 @@ vibebox explain     # 显示挂载与网络信息
 
 如果你想参与贡献 VibeBox，请先阅读 [贡献指南](CONTRIBUTING.md)，再提交 Pull Request。
 
-### 使用 VibeBox
-
-欢迎使用，也别忘了顺手帮 VibeBox 做点宣传！
-
 ### FAQ
 
 #### 它和其它 Sandboxes 有什么不同？
 
-Vibebox 追求的是：本地、可复现、启动快、流程简单。主要差异点：
+VibeBox 追求的是：本地、可复现、启动快、流程简单。主要差异点：
 
-- 在我的 M3 上，热启动通常 **6 秒以内**，可以非常快地回到工作状态。
+- 在我的 M3 上，热启动通常 **<5s**，可以非常快地回到工作状态。
 - 一个命令——`vibebox`——直接把你带进沙盒（从你的项目目录启动）。
 - 配置集中在 `vibebox.toml`，CPU / 内存 / 磁盘大小 / 挂载都能一眼看懂、随手改。
+- 会话是第一等公民：复用、多终端、清理。
 
 ### 特别鸣谢
 

docs/tasks.md

@@ -64,3 +64,4 @@
 3. [ ] Redirect vm output to log.
 4. [ ] Redirect vm output to vibebox starting it.
 5. [ ] use anyhow to sync api.
+6. [ ] add support for ipv6.

scripts/commit-check.sh

@@ -0,0 +1,6 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+cargo clippy --all-targets --all-features -- -D warnings
+cargo fmt --all
+cargo build --all-targets

src/bin/vibebox.rs

@@ -120,9 +120,15 @@ fn main() -> Result<()> {
     tracing::debug!(auto_shutdown_ms, "auto shutdown config");
     let manager_conn =
         vm_manager::ensure_manager(&raw_args, auto_shutdown_ms, config_override.as_deref())
-            .map_err(|err| color_eyre::eyre::eyre!(err.to_string()))?;
+            .map_err(|err| {
+                tracing::error!(error = %err, "failed to ensure vm manager");
+                color_eyre::eyre::eyre!(err.to_string())
+            })?;
 
-    instance::run_with_ssh(manager_conn).map_err(|err| color_eyre::eyre::eyre!(err.to_string()))?;
+    instance::run_with_ssh(manager_conn).map_err(|err| {
+        tracing::error!(error = %err, "failed to ensure vm manager");
+        color_eyre::eyre::eyre!(err.to_string())
+    })?;
 
     tracing::info!("See you again — keep vibecoding (no SEVs, only vibes) 😈");
 

src/instance.rs

@@ -21,6 +21,7 @@ use crate::{
 };
 
 const SSH_KEY_NAME: &str = "ssh_key";
+#[cfg_attr(feature = "mock-vm", allow(dead_code))]
 pub(crate) const VM_ROOT_LOG_NAME: &str = "vm_root.log";
 pub(crate) const STATUS_FILE_NAME: &str = "status.txt";
 pub(crate) const DEFAULT_SSH_USER: &str = "vibecoder";
@@ -68,7 +69,7 @@ pub fn run_with_ssh(manager_conn: UnixStream) -> Result<(), Box<dyn std::error::
     tracing::debug!(ssh_user = %ssh_user, "loaded instance config");
 
     let _manager_conn = manager_conn;
-    wait_for_vm_ipv4(&instance_dir, Duration::from_secs(120))?;
+    wait_for_vm_ipv4(&instance_dir, Duration::from_secs(480))?;
 
     let ip = load_or_create_instance_config(&instance_dir)?
         .vm_ipv4
@@ -217,6 +218,7 @@ fn generate_password() -> String {
     Uuid::now_v7().simple().to_string()
 }
 
+#[cfg_attr(feature = "mock-vm", allow(dead_code))]
 pub(crate) fn extract_ipv4(line: &str) -> Option<String> {
     let mut current = String::new();
     let mut best: Option<String> = None;
@@ -242,19 +244,46 @@ fn wait_for_vm_ipv4(
 ) -> Result<(), Box<dyn std::error::Error>> {
     let start = Instant::now();
     let mut next_log_at = start + Duration::from_secs(10);
+    let mut next_status_check = start;
     tracing::info!("waiting for vm ipv4");
     let status_path = instance_dir.join(STATUS_FILE_NAME);
     let mut last_status: Option<String> = None;
+    let mut status_missing = true;
     let mut once_hint = false;
     loop {
         let config = load_or_create_instance_config(instance_dir)?;
         if config.vm_ipv4.is_some() {
+            let _ = fs::remove_file(&status_path);
             return Ok(());
         }
         if start.elapsed() > timeout {
+            let _ = fs::remove_file(&status_path);
             return Err("Timed out waiting for VM IPv4".into());
         }
-        if Instant::now() >= next_log_at {
+        let now = Instant::now();
+        if now >= next_status_check {
+            match fs::read_to_string(&status_path) {
+                Ok(status) => {
+                    status_missing = false;
+                    let status = status.trim().to_string();
+                    if status.starts_with("error:") {
+                        let _ = fs::remove_file(&status_path);
+                        let message = status.trim_start_matches("error:").trim().to_string();
+                        return Err(message.into());
+                    }
+                    if !status.is_empty() && last_status.as_deref() != Some(status.as_str()) {
+                        tracing::info!("[background]: {}", status);
+                        last_status = Some(status);
+                        next_log_at = now + Duration::from_secs(20);
+                    }
+                }
+                Err(_) => {
+                    status_missing = true;
+                }
+            }
+            next_status_check = now + Duration::from_millis(500);
+        }
+        if now >= next_log_at {
             let waited = start.elapsed();
             if waited.as_secs() > 15 && !once_hint {
                 tracing::info!(
@@ -262,16 +291,10 @@ fn wait_for_vm_ipv4(
                 );
                 once_hint = true;
             }
-            if let Ok(status) = fs::read_to_string(&status_path) {
-                let status = status.trim().to_string();
-                if !status.is_empty() && last_status.as_deref() != Some(status.as_str()) {
-                    tracing::info!("[background]: {}", status);
-                    last_status = Some(status);
-                }
-            } else {
+            if status_missing {
                 tracing::info!("still waiting for vm ipv4, {}s elapsed", waited.as_secs(),);
             }
-            next_log_at += Duration::from_secs(10);
+            next_log_at += Duration::from_secs(20);
         }
         thread::sleep(Duration::from_millis(200));
     }
@@ -367,6 +390,7 @@ fn run_ssh_session(
     Ok(())
 }
 
+#[cfg_attr(feature = "mock-vm", allow(dead_code))]
 fn is_ipv4_candidate(candidate: &str) -> bool {
     let parts: Vec<&str> = candidate.split('.').collect();
     if parts.len() != 4 {

src/provision.sh

@@ -1,12 +1,55 @@
 #!/bin/bash
-set -eux
+set -eEux
+
+trap 'rc=$?; echo "[vibebox][error] provisioning failed at: ${BASH_COMMAND} (exit ${rc})"; echo "VIBEBOX_PROVISION_FAILED"; systemctl poweroff || true; exit 1' ERR
+
+# Wait for network + DNS before apt-get to avoid early boot flakiness.
+wait_for_network() {
+  echo "[vibebox] waiting for network/DNS readiness"
+  local deadline=$((SECONDS + 180))
+  while [ "$SECONDS" -lt "$deadline" ]; do
+    local has_route=0
+    if ip -4 route show default >/dev/null 2>&1; then
+      has_route=1
+    elif ip -6 route show default >/dev/null 2>&1; then
+      has_route=1
+    fi
+    if [ "$has_route" -eq 1 ]; then
+      if getent hosts deb.debian.org >/dev/null 2>&1; then
+        return 0
+      fi
+    fi
+    sleep 1
+  done
+  echo "[vibebox][warn] network/DNS still not ready after 180s; continuing" >&2
+  echo "[vibebox][warn] /etc/resolv.conf:" >&2
+  cat /etc/resolv.conf >&2 || true
+  ip -br addr >&2 || true
+  ip route >&2 || true
+  ip -6 route >&2 || true
+  return 0
+}
+
+apt_update_with_retries() {
+  local attempt=1
+  while [ "$attempt" -le 5 ]; do
+    if apt-get update; then
+      return 0
+    fi
+    echo "[vibebox][warn] apt-get update failed (attempt ${attempt}/5); retrying..." >&2
+    attempt=$((attempt + 1))
+    sleep 2
+  done
+  return 1
+}
 
 # Don't wait too long for slow mirrors.
-echo 'Acquire::http::Timeout "2";' | tee /etc/apt/apt.conf.d/99timeout
-echo 'Acquire::https::Timeout "2";' | tee -a /etc/apt/apt.conf.d/99timeout
-echo 'Acquire::Retries "2";' | tee -a /etc/apt/apt.conf.d/99timeout
+echo 'Acquire::http::Timeout "10";' | tee /etc/apt/apt.conf.d/99timeout
+echo 'Acquire::https::Timeout "10";' | tee -a /etc/apt/apt.conf.d/99timeout
+echo 'Acquire::Retries "5";' | tee -a /etc/apt/apt.conf.d/99timeout
 
-apt-get update
+wait_for_network
+apt_update_with_retries
 apt-get install -y --no-install-recommends      \
         build-essential                         \
         pkg-config                              \
@@ -55,4 +98,5 @@ sleep 100 # sleep here so that we don't see the login screen flash up before the
 EOF
 
 # Done provisioning, power off the VM
+echo "VIBEBOX_PROVISION_OK"
 systemctl poweroff

src/vm.rs

@@ -40,6 +40,7 @@ const DEFAULT_RAM_MB: u64 = 2048;
 const DEFAULT_RAM_BYTES: u64 = DEFAULT_RAM_MB * BYTES_PER_MB;
 const START_TIMEOUT: Duration = Duration::from_secs(60);
 const LOGIN_EXPECT_TIMEOUT: Duration = Duration::from_secs(120);
+const PROVISION_EXPECT_TIMEOUT: Duration = Duration::from_secs(900);
 
 struct StatusFile {
     path: PathBuf,
@@ -57,12 +58,6 @@ impl StatusFile {
     fn update(&self, message: &str) {
         let _ = fs::write(&self.path, message);
     }
-
-    fn clear(&self) {
-        if !self.cleared.swap(true, Ordering::SeqCst) {
-            let _ = fs::remove_file(&self.path);
-        }
-    }
 }
 
 impl Drop for StatusFile {
@@ -82,7 +77,15 @@ const BASE_DISK_RAW_NAME: &str = "disk.raw";
 
 #[derive(Clone)]
 pub(crate) enum LoginAction {
-    Expect { text: String, timeout: Duration },
+    Expect {
+        text: String,
+        timeout: Duration,
+    },
+    ExpectEither {
+        success: String,
+        failure: String,
+        timeout: Duration,
+    },
     Send(String),
 }
 use LoginAction::*;
@@ -365,6 +368,12 @@ enum WaitResult {
     Found,
 }
 
+#[derive(PartialEq, Eq)]
+enum WaitAnyResult {
+    Timeout,
+    Found(usize),
+}
+
 pub enum VmInput {
     Bytes(Vec<u8>),
     Shutdown,
@@ -372,6 +381,7 @@ pub enum VmInput {
 
 enum VmOutput {
     LoginActionTimeout { action: String, timeout: Duration },
+    LoginActionFailed { action: String, reason: String },
 }
 
 #[derive(Default)]
@@ -408,6 +418,41 @@ impl OutputMonitor {
             WaitResult::Found
         }
     }
+
+    fn wait_for_any(&self, needles: &[&str], timeout: Duration) -> WaitAnyResult {
+        let mut found: Option<usize> = None;
+        let (_unused, timeout_result) = self
+            .condvar
+            .wait_timeout_while(self.buffer.lock().unwrap(), timeout, |buf| {
+                if let Some((pos, idx, len)) = find_any(buf, needles) {
+                    *buf = buf[(pos + len)..].to_string();
+                    found = Some(idx);
+                    false
+                } else {
+                    true
+                }
+            })
+            .unwrap();
+
+        if timeout_result.timed_out() {
+            WaitAnyResult::Timeout
+        } else {
+            WaitAnyResult::Found(found.unwrap_or(0))
+        }
+    }
+}
+
+fn find_any(buf: &str, needles: &[&str]) -> Option<(usize, usize, usize)> {
+    let mut best: Option<(usize, usize, usize)> = None; // (pos, idx, len)
+    for (idx, needle) in needles.iter().enumerate() {
+        if let Some(pos) = buf.find(needle) {
+            let candidate = (pos, idx, needle.len());
+            if best.is_none_or(|b| candidate.0 < b.0) {
+                best = Some(candidate);
+            }
+        }
+    }
+    best
 }
 
 #[derive(Debug)]
@@ -552,14 +597,25 @@ fn ensure_default_image(
     fs::copy(base_raw, default_raw)?;
 
     let provision_command = script_command_from_content(PROVISION_SCRIPT_NAME, PROVISION_SCRIPT)?;
-    run_vm(
+    let provision_actions = [
+        Send(provision_command),
+        ExpectEither {
+            success: "VIBEBOX_PROVISION_OK".to_string(),
+            failure: "VIBEBOX_PROVISION_FAILED".to_string(),
+            timeout: PROVISION_EXPECT_TIMEOUT,
+        },
+    ];
+    if let Err(err) = run_vm(
         default_raw,
-        &[Send(provision_command)],
+        &provision_actions,
         directory_shares,
         DEFAULT_CPU_COUNT,
         DEFAULT_RAM_BYTES,
         None,
-    )?;
+    ) {
+        let _ = fs::remove_file(default_raw);
+        return Err(err);
+    }
 
     Ok(())
 }
@@ -1039,6 +1095,27 @@ fn spawn_login_actions_thread(
                         return;
                     }
                 }
+                ExpectEither {
+                    success,
+                    failure,
+                    timeout,
+                } => match output_monitor.wait_for_any(&[&success, &failure], timeout) {
+                    WaitAnyResult::Found(0) => {}
+                    WaitAnyResult::Found(_) => {
+                        let _ = vm_output_tx.send(VmOutput::LoginActionFailed {
+                            action: format!("expect '{}'", success),
+                            reason: format!("saw failure marker '{}'", failure),
+                        });
+                        return;
+                    }
+                    WaitAnyResult::Timeout => {
+                        let _ = vm_output_tx.send(VmOutput::LoginActionTimeout {
+                            action: format!("expect '{}'", success),
+                            timeout,
+                        });
+                        return;
+                    }
+                },
                 Send(mut text) => {
                     text.push('\n'); // Type the newline so the command is actually submitted.
                     input_tx.send(VmInput::Bytes(text.into_bytes())).unwrap();
@@ -1119,7 +1196,6 @@ where
 
     if let Some(status) = status {
         status.update("vm booting... go vibecoder!");
-        status.clear();
     }
     tracing::info!("vm booting");
 
@@ -1199,6 +1275,24 @@ where
                 }
                 break;
             }
+            Ok(VmOutput::LoginActionFailed { action, reason }) => {
+                exit_result = Err(format!(
+                    "Login action ({}) failed: {}; shutting down.",
+                    action, reason
+                )
+                .into());
+                unsafe {
+                    if vm.canRequestStop() {
+                        if let Err(err) = vm.requestStopWithError() {
+                            tracing::error!(error = ?err, "failed to request VM stop");
+                        }
+                    } else if vm.canStop() {
+                        let handler = RcBlock::new(|_error: *mut NSError| {});
+                        vm.stopWithCompletionHandler(&handler);
+                    }
+                }
+                break;
+            }
             Err(mpsc::TryRecvError::Empty) => {}
             Err(mpsc::TryRecvError::Disconnected) => {}
         }

src/vm_manager.rs

@@ -17,6 +17,7 @@ use std::{
 
 use crate::{
     config::CONFIG_PATH_ENV,
+    instance::STATUS_FILE_NAME,
     instance::VM_ROOT_LOG_NAME,
     instance::{
         DEFAULT_SSH_USER, InstanceConfig, build_ssh_login_actions, ensure_instance_dir,
@@ -30,8 +31,22 @@ use crate::{
 
 const VM_MANAGER_LOCK_NAME: &str = "vm.lock";
 const VM_MANAGER_LOG_NAME: &str = "vm_manager.log";
+const SHUTDOWN_RETRY_MS: u64 = 500;
+#[cfg(test)]
+const HARD_SHUTDOWN_TIMEOUT_MS: u64 = 1_000;
+#[cfg(not(test))]
 const HARD_SHUTDOWN_TIMEOUT_MS: u64 = 12_000;
 
+#[cfg(not(test))]
+fn force_exit(_reason: &str) -> ! {
+    std::process::exit(1);
+}
+
+#[cfg(test)]
+fn force_exit(reason: &str) -> ! {
+    panic!("{reason}");
+}
+
 pub fn ensure_manager(
     raw_args: &[std::ffi::OsString],
     auto_shutdown_ms: u64,
@@ -103,17 +118,37 @@ pub fn run_manager(
     let project_root = env::current_dir()?;
     tracing::info!(root = %project_root.display(), "vm manager starting");
     let _pid_guard = ensure_pid_file(&project_root)?;
-    run_manager_with(
-        &project_root,
-        args,
-        auto_shutdown_ms,
-        &RealVmExecutor,
-        ManagerOptions {
-            ensure_signed: true,
-            detach: true,
-            prepare_vm: true,
-        },
-    )
+    #[cfg(feature = "mock-vm")]
+    tracing::info!("vm manager using mock executor");
+    let executor: &dyn VmExecutor = {
+        #[cfg(feature = "mock-vm")]
+        {
+            &MockVmExecutor
+        }
+        #[cfg(not(feature = "mock-vm"))]
+        {
+            &RealVmExecutor
+        }
+    };
+    let options = {
+        #[cfg(feature = "mock-vm")]
+        {
+            ManagerOptions {
+                ensure_signed: false,
+                detach: true,
+                prepare_vm: false,
+            }
+        }
+        #[cfg(not(feature = "mock-vm"))]
+        {
+            ManagerOptions {
+                ensure_signed: true,
+                detach: true,
+                prepare_vm: true,
+            }
+        }
+    };
+    run_manager_with(&project_root, args, auto_shutdown_ms, executor, options)
 }
 
 fn spawn_manager_process(
@@ -468,6 +503,7 @@ fn read_client_pid(stream: &UnixStream) -> Option<u32> {
     }
 }
 
+#[cfg_attr(feature = "mock-vm", allow(dead_code))]
 fn spawn_manager_io(
     config: Arc<Mutex<InstanceConfig>>,
     instance_dir: PathBuf,
@@ -555,6 +591,7 @@ trait VmExecutor {
     ) -> Result<(), Box<dyn std::error::Error>>;
 }
 
+#[cfg_attr(feature = "mock-vm", allow(dead_code))]
 struct RealVmExecutor;
 
 impl VmExecutor for RealVmExecutor {
@@ -586,6 +623,39 @@ impl VmExecutor for RealVmExecutor {
     }
 }
 
+#[cfg(feature = "mock-vm")]
+struct MockVmExecutor;
+
+#[cfg(feature = "mock-vm")]
+impl VmExecutor for MockVmExecutor {
+    fn run_vm(
+        &self,
+        _args: vm::VmArg,
+        _extra_login_actions: Vec<LoginAction>,
+        _extra_shares: Vec<DirectoryShare>,
+        _config: Arc<Mutex<InstanceConfig>>,
+        _instance_dir: PathBuf,
+        vm_input_tx: Arc<Mutex<Option<mpsc::Sender<VmInput>>>>,
+    ) -> Result<(), Box<dyn std::error::Error>> {
+        let (tx, rx) = mpsc::channel::<VmInput>();
+        *vm_input_tx.lock().unwrap() = Some(tx);
+        tracing::info!("mock vm executor running");
+        while let Ok(input) = rx.recv() {
+            match input {
+                VmInput::Shutdown => break,
+                VmInput::Bytes(bytes) => {
+                    let text = String::from_utf8_lossy(&bytes);
+                    if text.contains("systemctl poweroff") {
+                        break;
+                    }
+                }
+            }
+        }
+        tracing::info!("mock vm executor exiting");
+        Ok(())
+    }
+}
+
 fn run_manager_with(
     project_root: &Path,
     mut args: vm::VmArg,
@@ -696,6 +766,10 @@ fn run_manager_with(
     );
     tracing::info!("vm manager vm run completed");
     let vm_err = vm_result.err().map(|e| e.to_string());
+    if let Some(err) = &vm_err {
+        let status_path = instance_dir.join(STATUS_FILE_NAME);
+        let _ = fs::write(&status_path, format!("error: {err}"));
+    }
     let _ = event_tx.send(ManagerEvent::VmExited(vm_err.clone()));
     let event_loop_result: Result<(), String> = event_loop_handle
         .join()
@@ -727,9 +801,14 @@ fn manager_event_loop(
     let hard_timeout = Duration::from_millis(HARD_SHUTDOWN_TIMEOUT_MS);
 
     loop {
-        let timeout = match shutdown_deadline {
-            Some(deadline) => deadline.saturating_duration_since(Instant::now()),
-            None => Duration::from_secs(1),
+        let timeout = match (shutdown_deadline, hard_deadline) {
+            (Some(shutdown), Some(hard)) => {
+                let next = if shutdown <= hard { shutdown } else { hard };
+                next.saturating_duration_since(Instant::now())
+            }
+            (Some(shutdown), None) => shutdown.saturating_duration_since(Instant::now()),
+            (None, Some(hard)) => hard.saturating_duration_since(Instant::now()),
+            (None, None) => Duration::from_secs(1),
         };
 
         match event_rx.recv_timeout(timeout) {
@@ -769,6 +848,9 @@ fn manager_event_loop(
                     && Instant::now() >= deadline
                     && !shutdown_sent
                 {
+                    if hard_deadline.is_none() {
+                        hard_deadline = Some(Instant::now() + hard_timeout);
+                    }
                     let mut sent = false;
                     if let Some(tx) = vm_input_tx.lock().unwrap().clone() {
                         if tx
@@ -788,19 +870,26 @@ fn manager_event_loop(
                         shutdown_deadline = None;
                         hard_deadline = Some(Instant::now() + hard_timeout);
                     } else {
-                        shutdown_deadline = Some(Instant::now() + Duration::from_millis(500));
+                        shutdown_deadline =
+                            Some(Instant::now() + Duration::from_millis(SHUTDOWN_RETRY_MS));
                     }
                 }
                 if ref_count == 0
-                    && shutdown_sent
                     && let Some(deadline) = hard_deadline
                     && Instant::now() >= deadline
                 {
-                    tracing::warn!(
-                        timeout_ms = HARD_SHUTDOWN_TIMEOUT_MS,
-                        "force exiting: VM did not stop after shutdown timeout"
-                    );
-                    std::process::exit(1);
+                    if shutdown_sent {
+                        tracing::warn!(
+                            timeout_ms = HARD_SHUTDOWN_TIMEOUT_MS,
+                            "force exiting: VM did not stop after shutdown timeout"
+                        );
+                    } else {
+                        tracing::warn!(
+                            timeout_ms = HARD_SHUTDOWN_TIMEOUT_MS,
+                            "force exiting: VM input not ready after shutdown timeout"
+                        );
+                    }
+                    force_exit("vm manager forced exit");
                 }
             }
             Err(mpsc::RecvTimeoutError::Disconnected) => break,
@@ -813,7 +902,7 @@ fn manager_event_loop(
 #[cfg(test)]
 mod tests {
     use super::*;
-    use std::{sync::mpsc, time::Duration};
+    use std::{sync::mpsc, thread, time::Duration};
 
     #[test]
     fn manager_powers_off_after_grace_when_no_refs() {
@@ -846,4 +935,53 @@ mod tests {
         let _ = event_tx.send(ManagerEvent::VmExited(None));
         let _ = manager_thread.join();
     }
+
+    #[test]
+    fn manager_force_exits_when_vm_input_never_ready() {
+        let (event_tx, event_rx) = mpsc::channel::<ManagerEvent>();
+        let vm_input_tx = Arc::new(Mutex::new(None));
+
+        let manager_thread = thread::spawn(move || {
+            let _ = manager_event_loop(event_rx, vm_input_tx, 10);
+        });
+
+        event_tx.send(ManagerEvent::Inc(None)).unwrap();
+        event_tx.send(ManagerEvent::Dec(None)).unwrap();
+
+        let join_result = manager_thread.join();
+        assert!(
+            join_result.is_err(),
+            "expected manager to force-exit when vm input never becomes ready"
+        );
+    }
+
+    #[test]
+    fn manager_sends_shutdown_after_vm_input_becomes_ready() {
+        let (event_tx, event_rx) = mpsc::channel::<ManagerEvent>();
+        let (vm_tx, vm_rx) = mpsc::channel::<VmInput>();
+        let vm_input_tx = Arc::new(Mutex::new(None));
+        let vm_input_for_thread = vm_input_tx.clone();
+
+        let manager_thread = thread::spawn(move || {
+            manager_event_loop(event_rx, vm_input_for_thread, 10).expect("event loop");
+        });
+
+        event_tx.send(ManagerEvent::Inc(None)).unwrap();
+        event_tx.send(ManagerEvent::Dec(None)).unwrap();
+
+        thread::sleep(Duration::from_millis(100));
+        *vm_input_tx.lock().unwrap() = Some(vm_tx);
+
+        let msg = vm_rx
+            .recv_timeout(Duration::from_secs(2))
+            .expect("poweroff");
+        match msg {
+            VmInput::Bytes(data) => {
+                assert_eq!(data, b"systemctl poweroff\n");
+            }
+            _ => panic!("unexpected vm input"),
+        }
+        let _ = event_tx.send(ManagerEvent::VmExited(None));
+        let _ = manager_thread.join();
+    }
 }

tests/e2e_cli.rs

@@ -0,0 +1,58 @@
+use assert_cmd::cargo::cargo_bin_cmd;
+use tempfile::TempDir;
+
+#[test]
+fn cli_version_shows_binary_name() {
+    let output = cargo_bin_cmd!("vibebox").arg("--version").output().unwrap();
+    print_output("e2e_cli", &output);
+    assert!(
+        output.status.success(),
+        "expected success, got status: {}",
+        output.status
+    );
+    let stdout = String::from_utf8_lossy(&output.stdout);
+    assert!(
+        stdout.contains("vibebox"),
+        "expected --version output to contain 'vibebox', got: {}",
+        stdout
+    );
+}
+
+#[test]
+fn list_reports_no_sessions_when_empty() {
+    let temp = TempDir::new().unwrap();
+    let home = temp.path().join("home");
+    let project = temp.path().join("project");
+    std::fs::create_dir_all(&home).unwrap();
+    std::fs::create_dir_all(&project).unwrap();
+
+    let output = cargo_bin_cmd!("vibebox")
+        .current_dir(&project)
+        .env("HOME", &home)
+        .arg("list")
+        .output()
+        .unwrap();
+    print_output("e2e_cli", &output);
+    assert!(
+        output.status.success(),
+        "expected success, got status: {}",
+        output.status
+    );
+    let stdout = String::from_utf8_lossy(&output.stdout);
+    assert!(
+        stdout.contains("No sessions were found."),
+        "expected empty sessions message, got: {}",
+        stdout
+    );
+}
+
+fn print_output(prefix: &str, output: &std::process::Output) {
+    let stdout = String::from_utf8_lossy(&output.stdout);
+    for line in stdout.lines() {
+        println!("[{}] {}", prefix, line);
+    }
+    let stderr = String::from_utf8_lossy(&output.stderr);
+    for line in stderr.lines() {
+        eprintln!("[{}] {}", prefix, line);
+    }
+}

tests/e2e_vm.rs

@@ -0,0 +1,278 @@
+use std::{
+    fs,
+    io::{BufRead, BufReader, Read},
+    path::{Path, PathBuf},
+    process::{Child, Command, Stdio},
+    thread,
+    time::{Duration, Instant},
+};
+
+use tempfile::TempDir;
+
+#[cfg(target_os = "macos")]
+#[test]
+#[ignore]
+fn vm_boots_and_runs_command() {
+    if std::env::var("VIBEBOX_E2E_VM").as_deref() != Ok("1") {
+        eprintln!("skipping: set VIBEBOX_E2E_VM=1 to run this test");
+        return;
+    }
+    if !virtualization_available() {
+        eprintln!("[e2e_vm] skipping: virtualization not available on this hardware");
+        return;
+    }
+
+    let temp = TempDir::new().unwrap();
+    let home = temp.path().join("home");
+    let cache_home = temp.path().join("cache");
+    let project = temp.path().join("project");
+    fs::create_dir_all(&home).unwrap();
+    fs::create_dir_all(&cache_home).unwrap();
+    fs::create_dir_all(&project).unwrap();
+
+    write_config(&project);
+
+    let mut child = Command::new(assert_cmd::cargo_bin!("vibebox-supervisor"))
+        .current_dir(&project)
+        .env("HOME", &home)
+        .env("XDG_CACHE_HOME", &cache_home)
+        .env("VIBEBOX_INTERNAL", "1")
+        .stdout(Stdio::piped())
+        .stderr(Stdio::piped())
+        .spawn()
+        .unwrap();
+    if let Some(stdout) = child.stdout.take() {
+        spawn_prefix_reader("e2e_vm", "stdout", stdout);
+    }
+    if let Some(stderr) = child.stderr.take() {
+        spawn_prefix_reader("e2e_vm", "stderr", stderr);
+    }
+    let _child_guard = ChildGuard::new(child);
+
+    log_line("e2e_vm", "waiting for vm manager socket");
+    let socket_path = project.join(".vibebox").join("vm.sock");
+    let _socket_guard = wait_for_socket(&socket_path, Duration::from_secs(30));
+    log_line("e2e_vm", "vm manager socket ready");
+
+    log_line("e2e_vm", "waiting for vm ipv4");
+    let instance_path = project.join(".vibebox").join("instance.toml");
+    let (ip, user) = wait_for_vm_ip(&instance_path, Duration::from_secs(180));
+    log_line("e2e_vm", &format!("vm ipv4={ip} user={user}"));
+
+    let ssh_key = project.join(".vibebox").join("ssh_key");
+    wait_for_file(&ssh_key, Duration::from_secs(30));
+    log_line("e2e_vm", "ssh key ready");
+
+    let output = wait_for_ssh_command(&ssh_key, &user, &ip, Duration::from_secs(90));
+    print_output("e2e_vm", &output);
+    let stdout = String::from_utf8_lossy(&output.stdout);
+    assert!(
+        stdout.contains("Linux"),
+        "expected ssh command output to contain 'Linux', got: {}",
+        stdout
+    );
+}
+
+#[cfg(not(target_os = "macos"))]
+#[test]
+#[ignore]
+fn vm_boots_and_runs_command() {
+    eprintln!("skipping: vm e2e test requires macOS virtualization");
+}
+
+struct ChildGuard {
+    child: Option<Child>,
+}
+
+impl ChildGuard {
+    fn new(child: Child) -> Self {
+        Self { child: Some(child) }
+    }
+}
+
+impl Drop for ChildGuard {
+    fn drop(&mut self) {
+        if let Some(child) = &mut self.child {
+            let _ = child.kill();
+            let _ = child.wait();
+        }
+    }
+}
+
+struct SocketGuard {
+    _path: PathBuf,
+    _stream: std::os::unix::net::UnixStream,
+}
+
+fn write_config(project: &Path) {
+    let config = r#"[box]
+cpu_count = 2
+ram_mb = 2048
+disk_gb = 5
+mounts = []
+
+[supervisor]
+auto_shutdown_ms = 120000
+"#;
+    fs::write(project.join("vibebox.toml"), config).unwrap();
+}
+
+fn wait_for_socket(path: &Path, timeout: Duration) -> SocketGuard {
+    let start = Instant::now();
+    loop {
+        if let Ok(stream) = std::os::unix::net::UnixStream::connect(path) {
+            return SocketGuard {
+                _path: path.to_path_buf(),
+                _stream: stream,
+            };
+        }
+        if start.elapsed() > timeout {
+            panic!(
+                "timed out waiting for vm manager socket at {}",
+                path.display()
+            );
+        }
+        thread::sleep(Duration::from_millis(200));
+    }
+}
+
+fn wait_for_vm_ip(instance_path: &Path, timeout: Duration) -> (String, String) {
+    let start = Instant::now();
+    loop {
+        if let Ok(raw) = fs::read_to_string(instance_path)
+            && let Ok(value) = toml::from_str::<toml::Value>(&raw)
+        {
+            let ip = value
+                .get("vm_ipv4")
+                .and_then(|v| v.as_str())
+                .map(|s| s.trim().to_string())
+                .filter(|s| !s.is_empty());
+            if let Some(ip) = ip {
+                let user = value
+                    .get("ssh_user")
+                    .and_then(|v| v.as_str())
+                    .map(|s| s.trim().to_string())
+                    .filter(|s| !s.is_empty())
+                    .unwrap_or_else(|| "vibecoder".to_string());
+                return (ip, user);
+            }
+        }
+        if start.elapsed() > timeout {
+            panic!(
+                "timed out waiting for vm_ipv4 in {}",
+                instance_path.display()
+            );
+        }
+        thread::sleep(Duration::from_millis(500));
+    }
+}
+
+fn wait_for_file(path: &Path, timeout: Duration) {
+    let start = Instant::now();
+    loop {
+        if path.exists() {
+            return;
+        }
+        if start.elapsed() > timeout {
+            panic!("timed out waiting for file {}", path.display());
+        }
+        thread::sleep(Duration::from_millis(200));
+    }
+}
+
+fn wait_for_ssh_command(
+    ssh_key: &Path,
+    user: &str,
+    ip: &str,
+    timeout: Duration,
+) -> std::process::Output {
+    let start = Instant::now();
+    loop {
+        let output = Command::new("ssh")
+            .args([
+                "-i",
+                ssh_key.to_str().unwrap_or(".vibebox/ssh_key"),
+                "-o",
+                "IdentitiesOnly=yes",
+                "-o",
+                "StrictHostKeyChecking=no",
+                "-o",
+                "UserKnownHostsFile=/dev/null",
+                "-o",
+                "GlobalKnownHostsFile=/dev/null",
+                "-o",
+                "PasswordAuthentication=no",
+                "-o",
+                "BatchMode=yes",
+                "-o",
+                "LogLevel=ERROR",
+                "-o",
+                "ConnectTimeout=5",
+                &format!("{user}@{ip}"),
+                "uname -s",
+            ])
+            .output()
+            .unwrap();
+        if output.status.success() {
+            return output;
+        }
+        if start.elapsed() > timeout {
+            let stderr = String::from_utf8_lossy(&output.stderr);
+            panic!("ssh command failed and timed out: {}", stderr);
+        }
+        thread::sleep(Duration::from_millis(1000));
+    }
+}
+
+fn spawn_prefix_reader(
+    label: &'static str,
+    stream: &'static str,
+    reader: impl Read + Send + 'static,
+) {
+    thread::spawn(move || {
+        let buf = BufReader::new(reader);
+        for line in buf.lines() {
+            match line {
+                Ok(line) => {
+                    println!("[{}][{}] {}", label, stream, line);
+                }
+                Err(err) => {
+                    eprintln!("[{}][{}] read error: {}", label, stream, err);
+                    break;
+                }
+            }
+        }
+    });
+}
+
+fn log_line(prefix: &str, message: &str) {
+    println!("[{}] {}", prefix, message);
+}
+
+fn print_output(prefix: &str, output: &std::process::Output) {
+    let stdout = String::from_utf8_lossy(&output.stdout);
+    for line in stdout.lines() {
+        println!("[{}] {}", prefix, line);
+    }
+    let stderr = String::from_utf8_lossy(&output.stderr);
+    for line in stderr.lines() {
+        eprintln!("[{}] {}", prefix, line);
+    }
+}
+
+fn virtualization_available() -> bool {
+    let output = Command::new("sysctl")
+        .args(["-n", "kern.hv_support"])
+        .output();
+    match output {
+        Ok(output) if output.status.success() => {
+            let value = String::from_utf8_lossy(&output.stdout);
+            match value.trim() {
+                "1" => true,
+                "0" => false,
+                _ => true,
+            }
+        }
+        _ => true,
+    }
+}

tests/e2e_vm_mock.rs

@@ -0,0 +1,455 @@
+#![cfg(all(feature = "mock-vm", target_os = "macos"))]
+
+use std::{
+    fs,
+    io::{BufRead, BufReader, Read},
+    os::unix::net::UnixStream,
+    path::{Path, PathBuf},
+    process::{Child, Command, Stdio},
+    sync::mpsc,
+    thread,
+    time::{Duration, Instant},
+};
+
+use tempfile::TempDir;
+
+#[test]
+fn mock_vm_allows_refcount_concurrency() {
+    let temp = TempDir::new().unwrap();
+    let mut supervisor = spawn_supervisor(&temp, 0, 1200, "e2e_vm_mock".to_string());
+
+    supervisor.clients = connect_clients(
+        &supervisor.socket_path,
+        12,
+        Duration::from_secs(2),
+        true,
+        "e2e_vm_mock",
+    );
+    log_line("e2e_vm_mock", "connected 12 clients");
+
+    assert_manager_alive_for(
+        &mut supervisor.child,
+        Duration::from_millis(900),
+        "vm manager exited while clients active",
+    );
+
+    let remaining = supervisor.clients.split_off(6);
+    supervisor.clients = remaining;
+    log_line("e2e_vm_mock", "dropped 6 clients");
+    assert_manager_alive_for(
+        &mut supervisor.child,
+        Duration::from_millis(900),
+        "vm manager exited while clients active",
+    );
+
+    supervisor.clients.clear();
+    log_line("e2e_vm_mock", "dropped remaining clients");
+    wait_for_exit(&mut supervisor.child, Duration::from_secs(10));
+    let status = supervisor.child.wait().unwrap();
+    assert!(status.success(), "vm manager exited with {status}");
+}
+
+#[test]
+fn mock_vm_many_managers_many_clients() {
+    let temp = TempDir::new().unwrap();
+    let mut supervisors = Vec::new();
+
+    for idx in 0..3 {
+        supervisors.push(spawn_supervisor(
+            &temp,
+            idx + 1,
+            1400,
+            format!("e2e_vm_mock_{idx}"),
+        ));
+    }
+
+    for supervisor in &mut supervisors {
+        supervisor.clients = connect_clients(
+            &supervisor.socket_path,
+            8,
+            Duration::from_secs(2),
+            true,
+            "e2e_vm_mock",
+        );
+    }
+    log_line("e2e_vm_mock", "connected 8 clients per manager");
+
+    for supervisor in &mut supervisors {
+        assert_manager_alive_for(
+            &mut supervisor.child,
+            Duration::from_millis(900),
+            "manager exited while clients active",
+        );
+    }
+
+    supervisors[0].clients.clear();
+    log_line("e2e_vm_mock", "dropped all clients for manager 0");
+    wait_for_exit(&mut supervisors[0].child, Duration::from_secs(10));
+    let status = supervisors[0].child.wait().unwrap();
+    assert!(status.success(), "manager 0 exited with {status}");
+
+    for supervisor in supervisors.iter_mut().skip(1) {
+        assert_manager_alive_for(
+            &mut supervisor.child,
+            Duration::from_millis(900),
+            "another manager exited early",
+        );
+    }
+
+    for supervisor in supervisors.iter_mut().skip(1) {
+        supervisor.clients.clear();
+    }
+    log_line("e2e_vm_mock", "dropped remaining clients");
+    for supervisor in supervisors.iter_mut().skip(1) {
+        wait_for_exit(&mut supervisor.child, Duration::from_secs(10));
+        let status = supervisor.child.wait().unwrap();
+        assert!(status.success(), "manager exited with {status}");
+    }
+}
+
+#[test]
+fn mock_vm_monkey_processes() {
+    let temp = TempDir::new().unwrap();
+    let mut rng = Lcg::new(0x5eed_f00d_dead_beef);
+    let mut supervisors = Vec::new();
+    let mut next_id = 0usize;
+    let max_supervisors = 4usize;
+    let steps = 25usize;
+
+    supervisors.push(spawn_supervisor(
+        &temp,
+        next_id,
+        1200,
+        format!("e2e_vm_monkey_{next_id}"),
+    ));
+    next_id += 1;
+
+    for step in 0..steps {
+        prune_exited_supervisors(&mut supervisors, "e2e_vm_monkey");
+        let roll = rng.gen_range(100);
+        log_line(
+            "e2e_vm_monkey",
+            &format!("step {step} roll={roll} supervisors={}", supervisors.len()),
+        );
+        if roll < 20 && supervisors.len() < max_supervisors {
+            supervisors.push(spawn_supervisor(
+                &temp,
+                next_id,
+                1200,
+                format!("e2e_vm_monkey_{next_id}"),
+            ));
+            log_line("e2e_vm_monkey", &format!("spawned supervisor {next_id}"));
+            next_id += 1;
+        } else if roll < 45 && !supervisors.is_empty() {
+            let idx = rng.gen_range(supervisors.len());
+            let mut supervisor = supervisors.swap_remove(idx);
+            log_line(
+                "e2e_vm_monkey",
+                &format!("killing supervisor {}", supervisor.label),
+            );
+            kill_supervisor(&mut supervisor, Duration::from_secs(5));
+        } else if roll < 80 && !supervisors.is_empty() {
+            let idx = rng.gen_range(supervisors.len());
+            let burst = 1 + rng.gen_range(3);
+            let new_clients = connect_clients(
+                &supervisors[idx].socket_path,
+                burst,
+                Duration::from_secs(1),
+                false,
+                "e2e_vm_monkey",
+            );
+            supervisors[idx].clients.extend(new_clients);
+            log_line(
+                "e2e_vm_monkey",
+                &format!("connected {burst} clients to {}", supervisors[idx].label),
+            );
+        } else if !supervisors.is_empty() {
+            let idx = rng.gen_range(supervisors.len());
+            if !supervisors[idx].clients.is_empty() {
+                let len = supervisors[idx].clients.len();
+                let drop_count = 1 + rng.gen_range(len);
+                supervisors[idx].clients.drain(0..drop_count.min(len));
+                log_line(
+                    "e2e_vm_monkey",
+                    &format!(
+                        "dropped {drop_count} clients from {}",
+                        supervisors[idx].label
+                    ),
+                );
+            }
+        }
+        thread::sleep(Duration::from_millis(200));
+    }
+
+    log_line("e2e_vm_monkey", "final cleanup");
+    for supervisor in supervisors.iter_mut() {
+        shutdown_supervisor(supervisor, Duration::from_secs(10));
+    }
+}
+
+#[test]
+fn mock_vm_exits_without_clients() {
+    let temp = TempDir::new().unwrap();
+    let mut supervisor = spawn_supervisor(&temp, 99, 300, "e2e_vm_no_clients".to_string());
+    wait_for_exit(&mut supervisor.child, Duration::from_secs(5));
+    let status = supervisor.child.wait().unwrap();
+    assert!(status.success(), "vm manager exited with {status}");
+}
+
+#[test]
+fn mock_vm_reconnect_resets_shutdown() {
+    let temp = TempDir::new().unwrap();
+    let mut supervisor = spawn_supervisor(&temp, 100, 800, "e2e_vm_reconnect".to_string());
+
+    supervisor.clients = connect_clients(
+        &supervisor.socket_path,
+        1,
+        Duration::from_secs(2),
+        true,
+        "e2e_vm_reconnect",
+    );
+    supervisor.clients.clear();
+    thread::sleep(Duration::from_millis(400));
+
+    supervisor.clients = connect_clients(
+        &supervisor.socket_path,
+        1,
+        Duration::from_secs(2),
+        true,
+        "e2e_vm_reconnect",
+    );
+    assert_manager_alive_for(
+        &mut supervisor.child,
+        Duration::from_millis(600),
+        "vm manager exited despite reconnect",
+    );
+
+    supervisor.clients.clear();
+    wait_for_exit(&mut supervisor.child, Duration::from_secs(10));
+    let status = supervisor.child.wait().unwrap();
+    assert!(status.success(), "vm manager exited with {status}");
+}
+
+struct Supervisor {
+    child: Child,
+    socket_path: PathBuf,
+    clients: Vec<UnixStream>,
+    label: String,
+}
+
+fn write_config(project: &Path, auto_shutdown_ms: u64) {
+    let config = format!(
+        r#"[box]
+cpu_count = 2
+ram_mb = 2048
+disk_gb = 5
+mounts = []
+
+[supervisor]
+auto_shutdown_ms = {auto_shutdown_ms}
+"#
+    );
+    fs::write(project.join("vibebox.toml"), config).unwrap();
+}
+
+fn spawn_supervisor(
+    temp: &TempDir,
+    idx: usize,
+    auto_shutdown_ms: u64,
+    label: String,
+) -> Supervisor {
+    let home = temp.path().join(format!("home-{idx}"));
+    let cache_home = temp.path().join(format!("cache-{idx}"));
+    let project = temp.path().join(format!("project-{idx}"));
+    fs::create_dir_all(&home).unwrap();
+    fs::create_dir_all(&cache_home).unwrap();
+    fs::create_dir_all(&project).unwrap();
+    write_config(&project, auto_shutdown_ms);
+
+    let mut child = Command::new(assert_cmd::cargo_bin!("vibebox-supervisor"))
+        .current_dir(&project)
+        .env("HOME", &home)
+        .env("XDG_CACHE_HOME", &cache_home)
+        .env("VIBEBOX_INTERNAL", "1")
+        .stdout(Stdio::piped())
+        .stderr(Stdio::piped())
+        .spawn()
+        .unwrap();
+    if let Some(stdout) = child.stdout.take() {
+        spawn_prefix_reader(label.clone(), "stdout", stdout);
+    }
+    if let Some(stderr) = child.stderr.take() {
+        spawn_prefix_reader(label.clone(), "stderr", stderr);
+    }
+
+    let socket_path = project.join(".vibebox").join("vm.sock");
+    wait_for_socket(&socket_path, Duration::from_secs(10));
+
+    Supervisor {
+        child,
+        socket_path,
+        clients: Vec::new(),
+        label,
+    }
+}
+
+fn connect_clients(
+    socket_path: &Path,
+    count: usize,
+    timeout: Duration,
+    require_all: bool,
+    label: &str,
+) -> Vec<UnixStream> {
+    let (tx, rx) = mpsc::channel();
+    let mut handles = Vec::with_capacity(count);
+    for _ in 0..count {
+        let path = socket_path.to_path_buf();
+        let tx = tx.clone();
+        handles.push(thread::spawn(move || {
+            let stream = connect_client_with_retry(&path, timeout);
+            let _ = tx.send(stream);
+        }));
+    }
+    drop(tx);
+    let mut clients = Vec::with_capacity(count);
+    for stream in rx.into_iter().flatten() {
+        clients.push(stream);
+    }
+    for handle in handles {
+        handle.join().unwrap();
+    }
+    if require_all && clients.len() != count {
+        panic!(
+            "client count mismatch: expected {count} got {}",
+            clients.len()
+        );
+    }
+    if !require_all && clients.len() != count {
+        log_line(
+            label,
+            &format!("connected {} of {count} clients", clients.len()),
+        );
+    }
+    clients
+}
+
+fn connect_client_with_retry(path: &Path, timeout: Duration) -> Option<UnixStream> {
+    let start = Instant::now();
+    loop {
+        match UnixStream::connect(path) {
+            Ok(stream) => return Some(stream),
+            Err(_) => {
+                if start.elapsed() > timeout {
+                    return None;
+                }
+                thread::sleep(Duration::from_millis(50));
+            }
+        }
+    }
+}
+
+fn wait_for_socket(path: &Path, timeout: Duration) {
+    let start = Instant::now();
+    loop {
+        if UnixStream::connect(path).is_ok() {
+            return;
+        }
+        if start.elapsed() > timeout {
+            panic!("timed out waiting for socket {}", path.display());
+        }
+        thread::sleep(Duration::from_millis(100));
+    }
+}
+
+fn assert_manager_alive(child: &mut Child, message: &str) {
+    assert!(child.try_wait().unwrap().is_none(), "{message}");
+}
+
+fn assert_manager_alive_for(child: &mut Child, duration: Duration, message: &str) {
+    let start = Instant::now();
+    while start.elapsed() < duration {
+        assert_manager_alive(child, message);
+        thread::sleep(Duration::from_millis(100));
+    }
+}
+
+fn kill_supervisor(supervisor: &mut Supervisor, timeout: Duration) {
+    supervisor.clients.clear();
+    let _ = supervisor.child.kill();
+    wait_for_exit(&mut supervisor.child, timeout);
+    let _ = supervisor.child.wait();
+}
+
+fn prune_exited_supervisors(supervisors: &mut Vec<Supervisor>, label: &str) {
+    supervisors.retain_mut(|supervisor| {
+        if supervisor.child.try_wait().unwrap().is_some() {
+            log_line(label, &format!("removed exited {}", supervisor.label));
+            false
+        } else {
+            true
+        }
+    });
+}
+
+fn shutdown_supervisor(supervisor: &mut Supervisor, timeout: Duration) {
+    supervisor.clients.clear();
+    wait_for_exit(&mut supervisor.child, timeout);
+    if supervisor.child.try_wait().unwrap().is_none() {
+        let _ = supervisor.child.kill();
+        let _ = supervisor.child.wait();
+    }
+}
+
+fn wait_for_exit(child: &mut Child, timeout: Duration) {
+    let start = Instant::now();
+    loop {
+        if child.try_wait().unwrap().is_some() {
+            return;
+        }
+        if start.elapsed() > timeout {
+            panic!("timed out waiting for mock vm supervisor exit");
+        }
+        thread::sleep(Duration::from_millis(200));
+    }
+}
+
+fn spawn_prefix_reader(label: String, stream: &'static str, reader: impl Read + Send + 'static) {
+    thread::spawn(move || {
+        let buf = BufReader::new(reader);
+        for line in buf.lines() {
+            match line {
+                Ok(line) => println!("[{}][{}] {}", label, stream, line),
+                Err(err) => {
+                    eprintln!("[{}][{}] read error: {}", label, stream, err);
+                    break;
+                }
+            }
+        }
+    });
+}
+
+fn log_line(prefix: &str, message: &str) {
+    println!("[{}] {}", prefix, message);
+}
+
+struct Lcg {
+    state: u64,
+}
+
+impl Lcg {
+    fn new(seed: u64) -> Self {
+        Self { state: seed }
+    }
+
+    fn next_u32(&mut self) -> u32 {
+        self.state = self.state.wrapping_mul(6364136223846793005).wrapping_add(1);
+        (self.state >> 32) as u32
+    }
+
+    fn gen_range(&mut self, upper: usize) -> usize {
+        if upper == 0 {
+            return 0;
+        }
+        (self.next_u32() as usize) % upper
+    }
+}

tests/explain_integration.rs

@@ -0,0 +1,113 @@
+use std::{env, ffi::OsString, fs, path::Path, sync::Mutex};
+
+use tempfile::TempDir;
+
+use vibebox::session_manager::INSTANCE_DIR_NAME;
+use vibebox::{config, explain};
+
+static ENV_MUTEX: Mutex<()> = Mutex::new(());
+
+struct EnvGuard {
+    key: &'static str,
+    previous: Option<OsString>,
+}
+
+impl EnvGuard {
+    fn set(key: &'static str, value: &Path) -> Self {
+        let previous = env::var_os(key);
+        unsafe {
+            env::set_var(key, value);
+        }
+        Self { key, previous }
+    }
+}
+
+impl Drop for EnvGuard {
+    fn drop(&mut self) {
+        match &self.previous {
+            Some(value) => unsafe {
+                env::set_var(self.key, value);
+            },
+            None => unsafe {
+                env::remove_var(self.key);
+            },
+        }
+    }
+}
+
+#[test]
+fn build_mount_rows_includes_defaults_and_custom_mounts() {
+    let _lock = ENV_MUTEX.lock().unwrap();
+    let temp = TempDir::new().unwrap();
+    let home = temp.path().join("home");
+    let project = home.join("project");
+    let cache_home = home.join("cache");
+    fs::create_dir_all(&project).unwrap();
+    fs::create_dir_all(&cache_home).unwrap();
+
+    let _home_guard = EnvGuard::set("HOME", &home);
+    let _cache_guard = EnvGuard::set("XDG_CACHE_HOME", &cache_home);
+
+    let box_cfg = config::BoxConfig {
+        mounts: vec!["data:~/data:read-only".to_string()],
+        ..Default::default()
+    };
+    let cfg = config::Config {
+        box_cfg,
+        supervisor: config::SupervisorConfig::default(),
+    };
+
+    let rows = explain::build_mount_rows(&project, &cfg).unwrap();
+
+    assert_eq!(rows.len(), 3);
+    assert_eq!(rows[0].host, "~/project");
+    assert_eq!(rows[0].guest, "~/project");
+    assert_eq!(rows[0].mode, "read-write");
+    assert_eq!(rows[0].default_mount, "yes");
+
+    assert_eq!(rows[1].host, "~/cache/vibebox/.guest-mise-cache");
+    assert_eq!(rows[1].guest, "/root/.local/share/mise");
+    assert_eq!(rows[1].mode, "read-write");
+    assert_eq!(rows[1].default_mount, "yes");
+
+    assert_eq!(rows[2].host, "~/project/data");
+    assert_eq!(rows[2].guest, "~/data");
+    assert_eq!(rows[2].mode, "read-only");
+    assert_eq!(rows[2].default_mount, "no");
+}
+
+#[test]
+fn build_network_rows_pending_without_instance_file() {
+    let temp = TempDir::new().unwrap();
+    let project = temp.path().join("project");
+    fs::create_dir_all(&project).unwrap();
+
+    let rows = explain::build_network_rows(&project).unwrap();
+
+    assert_eq!(rows.len(), 1);
+    assert_eq!(rows[0].network_type, "NAT");
+    assert_eq!(rows[0].vm_ip, "-");
+    assert_eq!(rows[0].host_to_vm, "ssh: <pending>:22");
+    assert_eq!(rows[0].vm_to_host, "none");
+}
+
+#[test]
+fn build_network_rows_uses_instance_vm_ip() {
+    let temp = TempDir::new().unwrap();
+    let project = temp.path().join("project");
+    let instance_dir = project.join(INSTANCE_DIR_NAME);
+    fs::create_dir_all(&instance_dir).unwrap();
+    fs::write(
+        instance_dir.join("instance.toml"),
+        "vm_ipv4 = \"10.1.2.3\"\n",
+    )
+    .unwrap();
+
+    let rows = explain::build_network_rows(&project).unwrap();
+
+    assert_eq!(rows.len(), 1);
+    assert_eq!(rows[0].network_type, "NAT");
+    assert_eq!(rows[0].vm_ip, "10.1.2.3");
+    assert_eq!(rows[0].host_to_vm, "ssh: 10.1.2.3:22");
+    assert_eq!(rows[0].vm_to_host, "none");
+}