Config Validator

验证和查询 OpenClaw 配置字段及有效值，提供配置示例并保证配置符合官方 schema 标准。

Security Scan

This is an instruction-only config-reference/validator skill that contains schema documents and guidance and does not request credentials, install software, or perform surprising actions—its declared purpose matches its contents.

Purpose & Capability

Name/description state: validate and query OpenClaw configuration. The package contains extensive schema markdown files and quick-reference material that directly support that purpose. It does not declare unrelated permissions or require unrelated binaries/credentials.

Instruction Scope

SKILL.md tells the agent to consult the included schema files to answer queries and validate configs; that is appropriate for an instruction-only validator. One operational note: the skill is instruction-only (no code), so automatic actions like running `openclaw doctor --fix` are described but not implemented by this skill itself — doing those requires the platform or user to run the commands. The SKILL.md's claim that it is automatically invoked by `safe-config-workflow` is plausible but depends on the platform orchestration (not provided by the skill).

Install Mechanism

No install spec, no downloads, and no code files to execute. Instruction-only skills have minimal installation risk.

Credentials

The skill declares no required environment variables or credentials, which is proportionate. The schema files do include example tokens/paths (e.g., a gateway auth.token value, serviceAccountFile path, proxy URL) but these are examples in documentation — they are not requests for secrets. Users should avoid copying real credentials into skill files or public registries.

Persistence & Privilege

always is false and there is no install behavior that modifies other skills or system-wide configuration. The skill does not request permanent presence or elevated privileges.

Assessment

This skill is essentially a local documentation + guidance pack for OpenClaw configuration and appears coherent with that goal. Before installing, verify the source/owner (currently unknown) and prefer using the skill only for reference: do not paste real private tokens or service-account JSON into the skill files or public locations. Remember the skill is instruction-only — it will answer queries based on the included markdown but it does not itself run `openclaw doctor` or apply fixes; running those commands requires your environment or platform to perform them. If you intend to have automated validation/fixes in workflows, confirm your platform's orchestration supports invoking this skill together with the real `openclaw` CLI in a safe, sandboxed context.

Config Validator Skill（配置验证器）

用途

验证 OpenClaw 配置字段和值的有效性，提供完整的 schema 参考和有效值枚举。

使用时机

✅ 推荐：修改配置前查阅（预防错误）

在思考如何修改配置时就先查阅本 Skill：

• 确认字段是否存在
• 确认有效值是什么
• 查看配置示例
• 避免写错配置

✅ 必须：修改配置后验证（兜底检查）

在safe-config-workflow 中自动调用验证：

• 验证配置是否正确
• doctor --fix 兜底修复

触发条件

当用户需要：

• 确认某个配置字段是否存在
• 查询某个字段的有效值
• 验证配置是否正确
• 查找配置示例
• 修改配置前参考（推荐）

模块结构

本 Skill 按模块组织配置 schema：

文件	内容
schema-channels.md	Channels 完整 schema（Telegram, WhatsApp, Discord, Slack...）
schema-agents.md	Agents 完整 schema（defaults, list, heartbeat, sandbox...）
schema-gateway.md	Gateway 完整 schema（port, bind, auth, reload...）
schema-session.md	Session 完整 schema（dmScope, identityLinks, reset...）
schema-tools.md	Tools 完整 schema（elevated, sandbox, allow/deny...）
schema-models.md	Models 完整 schema（providers, fallbacks, aliases...）
quick-reference.md	常用字段快速参考（50 个核心字段）

使用方法

1. 查询字段有效值

用户：telegram 的 streaming 字段有哪些有效值？
AI: 查阅 schema-channels.md → 返回有效值

2. 验证配置

用户：这个配置对吗？{ "channels": { "telegram": { "streaming": "on" } } }
AI: 查阅 schema-channels.md → 验证并反馈

3. 查找配置示例

用户：如何配置 Telegram 机器人？
AI: 查阅 schema-channels.md → 返回示例配置

验证规则

严格验证

• 字段必须存在于 schema 中
• 值必须是枚举中的有效值
• 类型必须匹配（string, number, boolean, object, array）

常见错误

错误类型	示例	正确值
无效枚举值	"streaming": "on"	"off" | "partial" | "block" | "progress"
无效类型	"port": "18789"	"port": 18789
未知字段	"channels": { "wechat": {} }	不支持的渠道

与 Safe Config Workflow 的集成

本 Skill 被 safe-config-workflow 自动调用：

safe-config-workflow 修改配置
    ↓
自动调用 config-validator 验证字段
    ↓
验证通过 → 继续执行
验证失败 → 反馈错误并停止

官方文档来源

所有 schema 信息来自官方文档：

• https://docs.openclaw.ai/gateway/configuration-reference
• https://docs.openclaw.ai/zh-CN/gateway/configuration

更新机制

当官方文档更新时：

1. 重新读取官方 configuration-reference.md
2. 更新对应的 schema 模块文件
3. 更新 quick-reference.md（如有变化）
4. 在 SKILL.md 中记录版本历史

版本历史

• v1.0 (2026-02-26) — 初始版本，包含完整 schema 和模块分文件