← 返回 Skills 市场

Config Validator

Name: Config Validator
Author: nicoxia

作者 nicoxia · GitHub ↗ · v1.0.1

cross-platform ✓ 安全检测通过

382

总下载

当前安装

版本数

在 OpenClaw 中安装

/install config-validator-zh-cn

功能描述

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。

使用说明 (SKILL.md)

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：

更新机制

当官方文档更新时：

重新读取官方 configuration-reference.md
更新对应的 schema 模块文件
更新 quick-reference.md（如有变化）
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

安全使用建议

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.

功能分析

Type: OpenClaw Skill Name: config-validator-zh-cn Version: 1.0.1 This OpenClaw skill bundle is a configuration validator and reference guide. All files are either metadata (`.json`) or documentation (`.md`) describing OpenClaw configuration schemas. The `SKILL.md` instructs the AI agent to 'consult' these documentation files to answer user queries or validate configurations, which is a read-only, informational task. There is no executable code, no prompt injection attempts against the agent, and no evidence of malicious intent such as data exfiltration, unauthorized execution, or persistence mechanisms. The documentation even details security-relevant configurations like tool sandboxing and command denial lists, indicating a focus on secure operation.

能力评估

✓ 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.

ℹ 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.

ℹ 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.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install config-validator-zh-cn
安装完成后，直接呼叫该 Skill 的名称或使用 /config-validator-zh-cn 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.1

GitHub 备份版本

v1.0.0

Config Validator Skill (配置验证器) v1.0.0 - 首次发布，支持验证 OpenClaw 配置字段和值的有效性。 - 提供完整 schema 模块分文件（channels、agents、gateway、session、tools、models、quick reference）。 - 支持字段存在性、有效值、类型匹配等严格验证。 - 集成于 safe-config-workflow，修改配置自动验证。 - 包含官方文档链接与字段快速参考。

元数据

Slug config-validator-zh-cn

版本 1.0.1

许可证 —

累计安装 0

当前安装数 0

历史版本数 2

常见问题