# Send Config Safe to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "config-safe",
    "name": "Config Safe",
    "source": "tencent",
    "type": "skill",
    "category": "安全合规",
    "sourceUrl": "https://clawhub.ai/glfruit/config-safe",
    "canonicalUrl": "https://clawhub.ai/glfruit/config-safe",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/config-safe",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=config-safe",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "config-safe",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-29T04:40:51.266Z",
      "expiresAt": "2026-05-06T04:40:51.266Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=config-safe",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=config-safe",
        "contentDisposition": "attachment; filename=\"config-safe-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "config-safe"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/config-safe"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/config-safe",
    "downloadUrl": "https://openagent3.xyz/downloads/config-safe",
    "agentUrl": "https://openagent3.xyz/skills/config-safe/agent",
    "manifestUrl": "https://openagent3.xyz/skills/config-safe/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/config-safe/agent.md"
  }
}
```
## Documentation

### Config Safe

安全地修改 OpenClaw 配置，避免写入无效配置导致 Gateway 无法启动。

### 核心理念

配置修改是高风险操作。

OpenClaw 有严格的配置验证，任何错误都会导致 Gateway 拒绝启动
一旦配置损坏，Gateway 崩溃，无法与你通信，只能通过 CLI 修复

核心原则：先预览，后验证，再确认，绝不直接修改。

### 验证方法

在修改配置之前，可以使用以下方法验证配置是否有效：

### 方法 1: Schema 验证（静态检查）

获取配置的 JSON Schema，检查字段类型和必填项：

openclaw gateway call config.schema --params '{}'

用途：

检查字段名是否正确
检查值的类型是否正确
检查必填字段是否遗漏

示例：

// Schema 返回结构
{
  "schema": {
    "type": "object",
    "properties": {
      "channels": { "type": "object" },
      "agents": { "type": "object" }
    },
    "required": ["agents"]
  }
}

### 方法 2: 模拟写入验证（动态检查）

使用 config.patch 尝试验证配置（不会实际写入，除非验证通过）：

openclaw gateway call config.patch --params '{
  "raw": "{\\"channels\\":{\\"telegram\\":{\\"enabled\\":true}}}",
  "baseHash": "<current-hash>"
}'

行为：

如果配置无效 → 返回错误，不写入，Gateway 继续运行
如果配置有效 → 写入并重启 Gateway

注意： 验证通过后会实际写入配置，所以要先向用户展示预览并获得确认。

### Step 1: 读取官方最新文档

在修改任何配置之前，先阅读相关文档：

# 配置总览和验证规则
cat /opt/homebrew/lib/node_modules/openclaw/docs/gateway/configuration.md

# 配置示例（常见场景）
cat /opt/homebrew/lib/node_modules/openclaw/docs/gateway/configuration-examples.md

必读内容：

严格配置验证 (Strict config validation)
配置字段的类型和默认值
相关功能的具体配置示例

### Step 2: 获取当前配置（只读）

openclaw gateway call config.get --params '{}'

只读取，不修改。 保存返回的 JSON 和 hash。

### Step 3: Schema 验证（可选但推荐）

先用 schema 验证配置结构：

openclaw gateway call config.schema --params '{}'

检查你的变更是否符合 schema 要求。

### Step 4: 生成预览

向用户展示变更内容：

=== 配置变更预览 ===

要修改:
- channels.telegram.enabled: false → true
- channels.telegram.botToken: [已隐藏]

变更前:
{
  "channels": { "telegram": { "enabled": false } }
}

变更后:
{
  "channels": { "telegram": { "enabled": true, "botToken": "***" } }
}

⚠️ 风险检查:
- 字段名正确 ✓
- 类型正确 ✓
- botToken 必填 ✓

### Step 5: 用户确认

必须明确获得用户确认后才继续：

请确认以上变更？输入 "确认" 继续，或 "取消" 放弃。

### Step 6: 验证并写入

# 部分更新（推荐）
openclaw gateway call config.patch --params '{
  "raw": "{\\"channels\\":{\\"telegram\\":{\\"enabled\\":true}}}",
  "baseHash": "<hash>"
}'

# 全量替换（仅当你完全理解风险时使用）
openclaw gateway call config.apply --params '{
  "raw": "<完整配置>",
  "baseHash": "<hash>"
}'

注意： 写入成功后 Gateway 会自动重启。

### Step 7: 验证结果

重启后检查配置是否生效：

openclaw status
openclaw doctor

### 添加/修改 channel

{
  channels: {
    telegram: {
      enabled: true,
      botToken: "your-token",
      dmPolicy: "pairing"
    }
  }
}

### 配置 agent 身份

{
  agents: {
    list: [{
      id: "main",
      identity: {
        name: "Samantha",
        emoji: "🦥"
      }
    }]
  }
}

### 配置 sandbox

{
  agents: {
    defaults: {
      sandbox: {
        mode: "all",
        scope: "session"
      }
    }
  }
}

### 错误处理

如果配置写入后 Gateway 无法启动：

运行 openclaw doctor 查看具体错误
运行 openclaw doctor --fix 尝试自动修复
如果无法修复，手动编辑 ~/.openclaw/openclaw.json

常见错误：

Unknown key: 字段名拼写错误
Invalid type: 值的类型不对
Missing required field: 缺少必填字段

### 黄金法则

在你这个技能中，永远不要：

❌ 直接写入配置而不预览
❌ 直接写入配置而不确认
❌ 使用 config.apply 而不提醒用户风险

你应该：

✅ 先读取文档
✅ Schema 验证配置结构
✅ 生成变更预览
✅ 明确要求用户确认
✅ 优先使用 config.patch

### 安全检查清单

在确认配置变更前，确认用户已经：

看到变更预览（变更前 vs 变更后）
 了解潜在风险
 输入 "确认" 明确同意

记住：Gateway 崩溃 = 通信中断 = 无法修复。预览 + 确认是唯一的防线。
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: glfruit
- Version: 1.0.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-04-29T04:40:51.266Z
- Expires at: 2026-05-06T04:40:51.266Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/config-safe)
- [Send to Agent page](https://openagent3.xyz/skills/config-safe/agent)
- [JSON manifest](https://openagent3.xyz/skills/config-safe/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/config-safe/agent.md)
- [Download page](https://openagent3.xyz/downloads/config-safe)