evex-agent-messaging
EVEX Agent Messaging Gateway
提供 Parent Main、Child Main 和 Role Child 对话操作的提供方中立 MCP 传输。Agent 调用此 MCP;它们不直接调用 OpenHands 的 Conversation API。
工具
create_child:验证完整的结构化 Mission 和干净的确定性 Git 检出,绑定 Child 身份及回调能力,然后创建一个确定性 Conversation。提供方运行一次无工具准入轮次,在 ACP 引导后恢复并重新验证确切的分支/head,然后才允许工具并交付 Mission。当 checkout 权限缺失或不匹配时,不会发生 Conversation API 调用。源角色仅接收 Messaging;只有明确的 QA/修复 Mission 才能请求capabilities: ["runtime_environment"]。send_to_parent:向能力所属的 Main 交付结构化的RESULT/NEEDS_INPUT。request_user_decision:向所属 Main 交付 A/B/C 式问题。cancel_mission/resume_mission:停止或恢复确切的 Child 任务。publish_navigation_links:向所属的 Main 发布信息性的 Issue/Main/Child/PR 链接。
能力是紧凑的不透明 evx1_ HMAC-SHA256 引用。它们绑定所属 Main、Child、任务键、角色、允许的操作和过期时间。服务器持有 OpenHands 凭据;它永远不会出现在工具输入、工具输出或子环境变量中。没有持久状态:调用方提供稳定的消息键,Main 对语义重放进行去重。
每个创建的 Child 都接收一个异步的原生 Stop 钩子。该钩子使用其作用域引用调用私有 /completion-hook 端点;网关等待终端原生状态,然后向所属 Main 发送一个包含有界终端助手响应的稳定 RECOVERY_WAKE。当模型在未调用 send_to_parent 的情况下完成时,这是提供方中立的回退。重复的钩子复用相同的语义消息键,不需要数据库、轮询器、读取工具或收据存储。
受信任的 Event Gateway/主机使用 main_capability_token(...) 铸造短期的 Main 能力,并将其注入 Main Mission。MCP 不向 Agent 暴露能力铸造工具。
Parent 仅创建或获取位于 /home/openhands/workspace/mirrors/<owner>--<repository>.git 的持久裸镜像。create_child 派生 /home/openhands/workspace/delivery/child-<child UUID>,在不存在时创建精确的分支工作树,并在接触 Conversation API 之前验证仓库、origin、分支、head 和干净状态。它从不远程克隆/获取、删除、重置或重新利用现有工作树。
Related MCP server: agent-broker
运行
服务器使用 MCP stdio 帧(Content-Length 头):
export EVEX_MESSAGING_SECRET='long-random-secret'
export OPENHANDS_URL='http://openhands:8000'
export OPENHANDS_API_KEY='server-only-key'
export OPENHANDS_PUBLIC_URL='http://openhands.example/canvas'
PYTHONPATH=src python3 -m evex_agent_messaging只有适配器读取 OpenHands 变量。在测试中或由嵌入该服务的主机可以注入假提供方,因此技能和 Mission 保持运行时中立。
验证
python3 -m unittest discover -s tests -p 'test_*.py' -qThis server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceMCP Server for AI agent identity and authorization. Create, verify, and manage agent identities with trust scores and scoped authorization tokens.MIT
- FlicenseNot gradedqualityCmaintenanceEnables multi-agent communication workflows with consensus arbitration, peer messaging, and operator-mediated collaboration through authenticated MCP tools.1
- AlicenseNot gradedqualityDmaintenanceEnables durable handoffs and shared scratchpad for multi-agent workflows over MCP HTTP transport.117MIT
- AlicenseNot gradedqualityBmaintenanceMCP server that enables AI agents on different machines to communicate and collaborate directly through relay channels, supporting structured agent contracts, real-time messaging, and human-in-the-loop approval workflows.1,905MIT
Related MCP Connectors
An authenticated remote MCP server for user-owned devices and one-shot capability invocation.
Workflow diagnostics, capability routing, and x402 settlement for MCP-compatible agents.
Agent-native collaboration network: orchestrate a team of long-running agents from any MCP client.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/EvexU2/evex-agent-messaging'
If you have feedback or need assistance with the MCP directory API, please join our Discord server