Add AgentBox to Execution Substrates & Sandboxing

README.md

@@ -2,9 +2,9 @@
 
 A curated, implementation-first list of **agent harness engineering** resources, with GitHub projects as the primary focus.
 
-- Total entries: **268**
-- GitHub entries: **241 (89.9%)**
-- GitHub in project categories (excluding readings): **236/236 (100.0%)**
+- Total entries: **269**
+- GitHub entries: **242 (90.0%)**
+- GitHub in project categories (excluding readings): **237/237 (100.0%)**
 - Categories: **9**
 - Last verified: **2026-06-05**
 - Language: [English](./README.md) | [中文](./README_zh.md)
@@ -48,7 +48,7 @@ A curated, implementation-first list of **agent harness engineering** resources,
 | --- | ---: |
 | Harness Architecture & Orchestration | 44 |
 | Context & Working-State Engineering | 16 |
-| Execution Substrates & Sandboxing | 25 |
+| Execution Substrates & Sandboxing | 26 |
 | Protocols, Tool Interfaces & Agent Contracts | 23 |
 | Evaluation Harnesses & Benchmarks | 27 |
 | Observability & Reliability Operations | 14 |
@@ -165,6 +165,7 @@ Notes:
 | agentbox | [GitHub](https://github.com/mattolson/agent-sandbox) | [![star](https://img.shields.io/badge/star-176-f4b400?style=flat-square)](https://github.com/mattolson/agent-sandbox) | sandbox, coding-agents, network-policy | Locked-down local sandbox for AI coding agents with scoped filesystem access, egress policy, secret injection, firewalling, and persistent agent state. |
 | HexAgent | [GitHub](https://github.com/UnicomAI/hexagent) | [![star](https://img.shields.io/badge/star-125-f4b400?style=flat-square)](https://github.com/UnicomAI/hexagent) | computer-layer, sandbox, runtime | Agent harness that separates the runtime from the computer it operates on through local, VM, and cloud sandbox backends. |
 | terminal-bench-env | [GitHub](https://github.com/ucsb-mlsec/terminal-bench-env) | [![star](https://img.shields.io/badge/star-83-f4b400?style=flat-square)](https://github.com/ucsb-mlsec/terminal-bench-env) | terminal, benchmark-env, sandbox | Environment layer for terminal-agent benchmark execution. |
+| AgentBox (madarco) | [GitHub](https://github.com/madarco/agentbox) | [![star](https://img.shields.io/badge/star-41-f4b400?style=flat-square)](https://github.com/madarco/agentbox) | sandbox, parallel, claude-code | CLI that runs coding agents in parallel, each teleported into its own sandboxed box (local Docker or cloud VMs) with checkpoints and a per-box browser/VS Code. |
 
 <a id="protocols-tool-interfaces-agent-contracts"></a>
 ### Protocols, Tool Interfaces & Agent Contracts

README_zh.md

@@ -2,9 +2,9 @@
 
 一个面向 **Agent Harness Engineering** 的工程实践清单，优先收录可直接落地的 GitHub 项目。
 
-- 当前条目数: **268**
-- GitHub 条目: **241 (89.9%)**
-- 项目分类 GitHub 占比（不含阅读类）: **236/236 (100.0%)**
+- 当前条目数: **269**
+- GitHub 条目: **242 (90.0%)**
+- 项目分类 GitHub 占比（不含阅读类）: **237/237 (100.0%)**
 - 分类数量: **9**
 - 最近核对日期: **2026-06-05**
 - 语言: [English](./README.md) | [中文](./README_zh.md)
@@ -48,7 +48,7 @@
 | --- | ---: |
 | Harness Architecture & Orchestration | 44 |
 | Context & Working-State Engineering | 16 |
-| Execution Substrates & Sandboxing | 25 |
+| Execution Substrates & Sandboxing | 26 |
 | Protocols, Tool Interfaces & Agent Contracts | 23 |
 | Evaluation Harnesses & Benchmarks | 27 |
 | Observability & Reliability Operations | 14 |
@@ -165,6 +165,7 @@
 | agentbox | [GitHub](https://github.com/mattolson/agent-sandbox) | [![star](https://img.shields.io/badge/star-176-f4b400?style=flat-square)](https://github.com/mattolson/agent-sandbox) | sandbox, coding-agents, network-policy | 面向 AI 编码代理的本地锁定沙箱，提供限定文件访问、出站策略、密钥注入、防火墙与持久代理状态。 |
 | HexAgent | [GitHub](https://github.com/UnicomAI/hexagent) | [![star](https://img.shields.io/badge/star-125-f4b400?style=flat-square)](https://github.com/UnicomAI/hexagent) | computer-layer, sandbox, runtime | 将代理运行时与其操作的计算机分离的 agent harness，支持本地、VM 与云端沙箱后端。 |
 | terminal-bench-env | [GitHub](https://github.com/ucsb-mlsec/terminal-bench-env) | [![star](https://img.shields.io/badge/star-83-f4b400?style=flat-square)](https://github.com/ucsb-mlsec/terminal-bench-env) | terminal, benchmark-env, sandbox | 为终端代理基准测试提供执行环境层。 |
+| AgentBox (madarco) | [GitHub](https://github.com/madarco/agentbox) | [![star](https://img.shields.io/badge/star-41-f4b400?style=flat-square)](https://github.com/madarco/agentbox) | sandbox, parallel, claude-code | 在并行的独立沙箱（本地 Docker 或云端虚拟机）中运行多个编码代理的 CLI，支持检查点以及每个盒子内置的浏览器/VS Code。 |
 
 <a id="protocols-tool-interfaces-agent-contracts"></a>
 ### Protocols, Tool Interfaces & Agent Contracts

data/projects.yaml

@@ -3725,3 +3725,16 @@ entries:
   updated_at: '2025-01-01'
   license: n/a
   why_included: High-signal practitioner framing for harness-first implementation.
+- name: AgentBox (madarco)
+  repo_url: https://github.com/madarco/agentbox
+  category: Execution Substrates & Sandboxing
+  summary_en: CLI that runs coding agents in parallel, each teleported into its own sandboxed box (local Docker or cloud VMs) with checkpoints and a per-box browser/VS Code.
+  summary_zh: 在并行的独立沙箱（本地 Docker 或云端虚拟机）中运行多个编码代理的 CLI，支持检查点以及每个盒子内置的浏览器/VS Code。
+  tags:
+  - sandbox
+  - parallel
+  - claude-code
+  stars_snapshot: 41
+  updated_at: '2026-06-06'
+  license: MIT
+  why_included: Self-hostable execution substrate that teleports a project into an isolated box per agent (local Docker FUSE overlay or Hetzner/Daytona/Vercel/E2B), with sub-1s checkpoint starts and host-held git credentials; works with Claude Code, Codex, and OpenCode.

reports/verification/2026-06-06.md

@@ -0,0 +1,67 @@
+# Verification Report
+
+- Generated at: `2026-06-06T08:50:17.872197+00:00`
+- Total entries: `269`
+- GitHub entries: `242` (90.0%)
+- GitHub in project categories (excluding `Essential Readings & Ecosystem Maps`): `237/237` (100.0%)
+- Categories: `9`
+- URL checks: `270` total, `270` reachable, `0` broken
+
+## Category Counts
+
+| Category | Entries |
+| --- | ---: |
+| Harness Architecture & Orchestration | 44 |
+| Context & Working-State Engineering | 16 |
+| Execution Substrates & Sandboxing | 26 |
+| Protocols, Tool Interfaces & Agent Contracts | 23 |
+| Evaluation Harnesses & Benchmarks | 27 |
+| Observability & Reliability Operations | 14 |
+| Guardrails, Security & Governance | 19 |
+| Reference Harness Implementations | 68 |
+| Essential Readings & Ecosystem Maps | 32 |
+
+## Structural Errors
+
+- None
+
+## Warnings
+
+- None
+
+## Broken URLs
+
+- None
+
+## Reachable URL Sample
+
+- `HEAD 200` https://blog.langchain.com/agent-frameworks-runtimes-and-harnesses-oh-my/
+- `HEAD 200` https://blog.langchain.com/evaluating-deep-agents-our-learnings/
+- `HEAD 200` https://blog.langchain.com/improving-deep-agents-with-harness-engineering/
+- `HEAD 200` https://blog.langchain.com/the-anatomy-of-an-agent-harness/
+- `HEAD 200` https://claude.com/blog/building-agents-with-the-claude-agent-sdk
+- `HEAD 200` https://cognition.ai/blog/what-we-learned-building-cloud-agents
+- `HEAD 200` https://developers.openai.com/blog/eval-skills
+- `HEAD 200` https://github.com/1jehuang/jcode
+- `HEAD 200` https://github.com/21st-dev/1code
+- `HEAD 200` https://github.com/2FastLabs/agent-squad
+- `HEAD 200` https://github.com/AVIDS2/memorix
+- `HEAD 200` https://github.com/AgentOps-AI/agentops
+- `HEAD 200` https://github.com/Aider-AI/aider
+- `HEAD 200` https://github.com/AndyMik90/Aperant
+- `HEAD 200` https://github.com/Arize-ai/openinference
+- `HEAD 200` https://github.com/Arize-ai/phoenix
+- `HEAD 200` https://github.com/Atmosphere/atmosphere
+- `HEAD 200` https://github.com/BerriAI/litellm
+- `HEAD 200` https://github.com/BloopAI/vibe-kanban
+- `HEAD 200` https://github.com/Chorus-AIDLC/Chorus
+- `HEAD 200` https://github.com/ChromeDevTools/chrome-devtools-mcp
+- `HEAD 200` https://github.com/ComposioHQ/agent-orchestrator
+- `HEAD 200` https://github.com/DevAgentForge/Open-Claude-Cowork
+- `HEAD 200` https://github.com/EleutherAI/lm-evaluation-harness
+- `HEAD 200` https://github.com/EveryInc/compound-engineering-plugin
+- `HEAD 200` https://github.com/FoundationAgents/OpenManus
+- `HEAD 200` https://github.com/Git-on-my-level/codex-autorunner
+- `HEAD 200` https://github.com/GoogleCloudPlatform/scion
+- `HEAD 200` https://github.com/HKUDS/CLI-Anything
+- `HEAD 200` https://github.com/HKUDS/OpenHarness