项目与实践

OpenCompany 用 company.toml、Agent Team 和 Human Sign-off 把组织结构与 Agent 配置写成可版本控制文件，并明确保留人工签字点。

Munim Computer Use reads the accessibility tree and operates desktop controls by element ID, reducing reliance on fragile screenshot-coordinate automation.

介绍 Munim Computer Use 如何先读取可访问性树，再按元素 ID 操作桌面控件，减少依赖截图坐标的脆弱自动化。

How jcode handles multi-agent resources, state synchronization, and verification in its harness, beyond the capabilities of the model alone.

介绍 jcode 如何把多 Agent 的资源、状态同步和验证放到 Harness 层，说明并行 Coding Agent 的瓶颈不只是模型能力。

介绍 GitHub Agentic Workflows 如何把需要判断的自动化放进 GitHub Actions，同时保留权限、输出、审批和安全边界。

How GitHub Agentic Workflows puts judgment-based automation inside GitHub Actions while retaining permissions, controlled outputs, approval, and security.

A multi-agent engineering workflow uses isolated worktrees, CI, cold-start review, evidence gates, and layered context to control delivery of AI-generated code.

介绍一个多 Agent 工程工作流，如何用独立 worktree、CI、冷启动审查、证据门和分层上下文控制 AI 生成代码的交付。

VEP/1 turns images into short evidence packets. Provider fallback, caching, and credential storage manage cost and reliability for text agents.

通过 VEP/1 把图片转换成短证据包，并用 Provider fallback、缓存和凭据存储控制文本 Agent 的视觉输入成本与可靠性。

A Hermes MiniMax plugin with image-01, Hailuo async video tasks, CN/global endpoint fallback, end-to-end tests, and a PyPI release.