openclaw-docs-online

OpenClaw 官方文档查询技能 - 指导如何访问和查询 https://docs.openclaw.ai/ 官方文档

Security Scan

This is an instruction-only skill to query the official OpenClaw docs (docs.openclaw.ai); it requests no credentials, no installs, and its requirements match its stated purpose.

Purpose & Capability

The name/description say it's a documentation lookup helper for https://docs.openclaw.ai/ and the SKILL.md/CLAUDE.md only instruct the agent to find and cite official documentation. There are no unrelated env vars, binaries, or install steps required — the requested capabilities are proportionate to a docs lookup skill.

Instruction Scope

The runtime instructions focus on web-based document lookup, structured extraction, and citing sources — all within scope. The examples include local commands and paths (e.g., ~/.openclaw/gateway, openclaw restart, viewing logs), which are plausible examples for troubleshooting but could be misinterpreted by an agent with shell access as actions to execute. The instructions do not explicitly tell the agent to read arbitrary system files or exfiltrate data.

Install Mechanism

No install spec or code is provided (instruction-only). Nothing is downloaded or written to disk by the skill itself, so install-risk is minimal.

Credentials

The skill declares no required environment variables, no credentials, and no config paths. It does note that network access is required to reach the official docs, which is expected for its purpose.

Persistence & Privilege

Flags show always:false and normal model invocation allowed. The skill does not request permanent presence nor modify other skills or system-wide settings. Autonomous invocation is the platform default and is not a problem here by itself.

Assessment

This skill appears coherent and matches its description: it helps the agent look up and cite https://docs.openclaw.ai/. Before installing, confirm whether you want the agent to have network access to fetch docs. Also review your agent's runtime privileges: if the agent is allowed to run shell commands or access local files, be aware the skill's examples reference local paths and restart commands (e.g., ~/.openclaw/gateway, openclaw restart); consider restricting shell/file access or requiring explicit user confirmation before the agent executes any local commands. If future versions add installs, external download URLs, or requests for API keys/credentials, re-evaluate (those would raise concerns).

OpenClaw 官方文档查询技能

指导 AI 如何访问和查询 OpenClaw 官方文档 (https://docs.openclaw.ai/)，获取准确、权威的产品信息。

核心功能

• 文档访问指导 - 指导如何正确访问官方文档
• 信息查询策略 - 提供高效的文档查询方法
• 内容验证机制 - 确保获取信息的准确性
• 结构化输出 - 提供清晰的查询结果

五要素框架

本技能严格遵循五要素框架设计：

WHEN（触发条件）

• 用户询问 OpenClaw 功能、配置、使用方法时
• 需要确认 OpenClaw 官方特性或限制时
• 遇到 OpenClaw 相关问题需要查找官方解答时
• 需要了解 OpenClaw 最新版本更新内容时

WHAT（具体成果）

• 输出基于官方文档的准确答案
• 提供文档来源链接和引用位置
• 给出可操作的具体步骤
• 标注信息的时效性和适用版本

HOW（步骤逻辑）

1. 识别查询意图，确定关键词
2. 访问官方文档，定位相关章节
3. 提取核心信息，验证准确性
4. 结构化输出，附带引用来源

REFERENCE（正反样例）

• 【正例】用户问"如何配置 Gateway"，访问文档后输出完整配置步骤和参数说明
• 【反例】凭记忆回答配置方法，未验证文档，导致参数错误

LIMITS（明确限制）

• 必须访问官方文档验证信息
• 不得凭记忆或推测回答技术问题
• 必须标注信息来源和时效性
• 文档无法找到答案时必须明确告知

---

新技能已创建成功！ 🎉

📝 注意：

• 本技能需要网络访问能力
• 查询结果以官方文档为准
• 建议定期更新文档知识