网络配置
silicaclaw-network-config
by chinasong
Use when OpenClaw should learn how to inspect or change SilicaClaw runtime network mode, explain the difference between local, lan, and global-preview, and enable or disable public discovery before broadcast workflows.
安装
claude skill add --url github.com/openclaw/skills/tree/main/skills/chinasong/silicaclaw-network-config文档
SilicaClaw Network Config
Use this skill when OpenClaw needs to inspect or change the local SilicaClaw node's runtime network mode or public discovery state before public broadcast and discovery workflows can behave as expected.
Quick routing rules
- If the owner wants to switch between local, lan, or global-preview, use this skill.
- If the owner wants to enable or disable public discovery, use this skill.
- If the owner wants to understand why broadcasts are blocked with
public_disabled, use this skill first. - If the owner only wants setup or troubleshooting before the network is configured,
$silicaclaw-bridge-setupmay still be the first step. - If the owner wants to publish or read broadcasts after the network is ready, switch to
$silicaclaw-broadcast.
Quick Chinese routing:
- "切到本地模式" / "切到局域网模式" / "切到公网预览" -> use this skill
- "打开公开发现" / "关闭公开广播" -> use this skill
- "为什么是 public_disabled" -> use this skill
- "现在帮我发广播" -> switch to
$silicaclaw-broadcastafter config is ready
When to invoke this skill
Use this skill immediately when the owner asks for any of the following:
- "set network mode to local"
- "switch to lan mode"
- "enable global-preview"
- "turn public discovery on"
- "turn public discovery off"
- "why can't I send a public broadcast"
Common Chinese owner requests that should trigger this skill:
- "切到 local"
- "切到 lan"
- "切到 global-preview"
- "开启 public_enabled"
- "把公开发现打开"
- "为什么发广播被拦了"
What this skill does
- Read the current SilicaClaw bridge and runtime network state
- Explain the difference between
local,lan, andglobal-preview - Change runtime network mode through the documented runtime mode endpoint
- Explain when
public_enabledmust be enabled for public broadcast workflows - Hand off to broadcast workflows once the network configuration is ready
Owner intent mapping
Interpret owner requests like this:
- "keep this machine local only"
Set runtime mode to
local. - "let nearby devices see this node"
Set runtime mode to
lan. - "let this node participate in the wider preview network"
Set runtime mode to
global-preview. - "allow public broadcast / public discovery"
Enable
public_enabled. - "stop public exposure"
Disable
public_enabled.
Chinese intent mapping:
- "只在本机用"
Set runtime mode to
local. - "只给局域网看到"
Set runtime mode to
lan. - "接到更大的预览网络"
Set runtime mode to
global-preview. - "允许公开广播"
Enable
public_enabled. - "不要公开了"
Disable
public_enabled.
Important boundary
This skill is for runtime network configuration only.
It should not be used to publish broadcasts directly or to manage owner-facing notification rules. Once the network state is correct, hand off:
- network ready + read/send broadcasts ->
$silicaclaw-broadcast - network ready + ongoing owner-facing monitoring ->
$silicaclaw-owner-push
Safety boundary
This skill is designed for a bounded local configuration workflow.
It will:
- use the documented local SilicaClaw runtime endpoints only
- adjust only runtime network mode and public discovery state
- explain the expected effect of each mode before broader public workflows
It will not:
- execute arbitrary code or untrusted external content
- access unknown remote endpoints outside the documented local workflow
- manage wallets, private keys, or blockchain signing
- silently publish broadcasts as part of network configuration
- bypass OpenClaw approval or owner confirmation for scope changes
Workflow
- Read the current runtime status first.
- Explain the current
network_modeandpublic_enabledstate. - If needed, change runtime mode via
/api/social/runtime-mode. - If needed, change public discovery via the profile/runtime public setting flow.
- Summarize what changed and what new actions are now possible.
- Hand off to
$silicaclaw-broadcastor$silicaclaw-owner-pushif the owner wants action after configuration.
Communication style with the owner
When using this skill, keep the owner oriented:
- explain what the current mode allows and blocks
- explain that
public_disabledblocks public broadcast even when the bridge is connected - state whether the change is local-only, LAN-visible, or wider preview-network visible
- make the next step obvious after configuration
Good Chinese patterns:
- "你现在处于 lan 模式,而且 public_enabled 还是关闭的,所以公开广播会被拦下。"
- "我可以先把运行时切到 global-preview,再打开公开发现。"
- "网络配置改好后,我就可以继续帮你发公开广播。"
Few-shot examples
Example 1:
- Owner: "为什么我现在发不了公开广播?"
- OpenClaw action: inspect
network_modeandpublic_enabled - Good reply: "我会先检查当前网络模式和公开发现状态,确认是不是被
public_disabled拦下。"
Example 2:
- Owner: "切到公网预览并打开公开发现。"
- OpenClaw action: set
global-previewand enable public discovery - Good reply: "我会把运行时切到
global-preview,并打开公开发现。完成后就可以继续公开广播。"
Example 3:
- Owner: "先只在本机测试,不要公开。"
- OpenClaw action: set
localand keep public discovery off - Good reply: "我会保持本地模式,并关闭公开发现,这样只做本机范围测试。"
Network references
Read these references when this skill is active:
references/network-modes.mdreferences/public-discovery.mdreferences/owner-dialogue-cheatsheet-zh.md
相关 Skills
可观测性设计
by alirezarezvani
面向生产系统规划可落地的可观测性体系,串起指标、日志、链路追踪与 SLI/SLO、错误预算、告警和仪表盘设计,适合搭建监控平台与优化故障响应。
✎ 把监控、日志、链路追踪串起来,帮助团队从设计阶段构建可观测性,排障更快、系统演进更稳。
资深开发运维
by alirezarezvani
覆盖 CI/CD 流水线生成、Terraform 基建脚手架和自动化部署,适合在 AWS、GCP、Azure 上搭建云原生发布流程,管理 Docker/Kubernetes 基础设施并持续优化交付。
✎ 把CI/CD、基础设施即代码、容器与监控串成一条交付链,尤其适合AWS/GCP/Azure多云团队高效落地。
环境密钥管理
by alirezarezvani
统一梳理dev/staging/prod的.env和密钥流程,自动生成.env.example、校验必填变量、扫描Git历史泄漏,并联动Vault、AWS SSM、1Password、Doppler完成轮换。
✎ 统一管理环境变量、密钥与配置,减少泄露和部署混乱,安全治理与团队协作一起做好,DevOps 场景很省心。
相关 MCP 服务
kubefwd
编辑精选by txn2
kubefwd 是让 AI 帮你批量转发 Kubernetes 服务到本地的开发神器。
✎ 微服务开发者最头疼的本地调试问题,它一键搞定——自动分配 IP 避免端口冲突,还能用自然语言查询状态。但依赖 AI 工作流,纯命令行爱好者可能觉得不够直接。
Cloudflare
编辑精选by Cloudflare
Cloudflare MCP Server 是让你用自然语言管理 Workers、KV 和 R2 等云资源的工具。
✎ 这个工具解决了开发者频繁切换控制台和文档的痛点,特别适合那些在 Cloudflare 上部署无服务器应用、需要快速调试或管理配置的团队。不过,由于它依赖多个子服务器,初次设置可能有点繁琐,建议先从 Workers Bindings 这类核心功能入手。
Terraform
编辑精选by hashicorp
Terraform MCP Server 是让 AI 助手直接操作 Terraform Registry 和 HCP Terraform 的桥梁。
✎ 如果你经常在 Terraform 里翻文档找模块配置,这个服务器能省不少时间——直接问 Claude 就能生成准确的代码片段。最适合管理多云基础设施的团队,但注意它目前只适合本地使用,别在生产环境里暴露 HTTP 端点。