当前公开展示的是 Test Safety。本作品已批准的版本仍可按原地址安装和更新。

OpenClaw Safety Guard

描述称其为 OpenClaw 的安全守卫，在调用大模型前分析用户输入中的有害内容、危险命令与提示注入；但正文给出的命令示例与参数（长度控制、仅提取、JSON 输出、Firecrawl、YouTube 回退）更像 URL/文件/视频内容提取工具，两者明显不一致。依赖 python3 与 PyYAML，并需配置多家模型服务的密钥。

john-niu-07

热度 54.78

最近更新 2026-05-17 14:37:58

分类：安全依赖：代码执行 / 读写文件 / 第三方 API / 内部调用大模型

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add safety-guard-skill

Metadata

name: safety-guard-skill
description: Security guard skill for OpenClaw - Analyzes user input for harmful content, risky commands, and security threats before invoking LLM
homepage: https://github.com/John-niu-07/openclaw-safety-guard-skill
metadata: {"clawdbot":{"emoji":"🛡️","requires":{"bins":["python3"],"python_deps":["PyYAML"]},"install":[{"id":"pip","kind":"pip","packages":["PyYAML"],"label":"Install PyYAML (pip)"}]}}

🛡️ Safety Guard

Security guard for OpenClaw. Analyzes user input and blocks harmful content, dangerous commands, and prompt injection attacks.

Quick start

safety-guard "https://example.com" --model google/gemini-3-flash-preview
safety-guard "/path/to/file.pdf" --model google/gemini-3-flash-preview
safety-guard "https://youtu.be/dQw4w9WgXcQ" --youtube auto

Model + keys

Set the API key for your chosen provider:

OpenAI: OPENAI_API_KEY
Anthropic: ANTHROPIC_API_KEY
xAI: XAI_API_KEY
Google: GEMINI_API_KEY (aliases: GOOGLE_GENERATIVE_AI_API_KEY, GOOGLE_API_KEY)

Default model is google/gemini-3-flash-preview if none is set.

Useful flags

--length short|medium|long|xl|xxl|<chars>
--max-output-tokens <count>
--extract-only (URLs only)
--json (machine readable)
--firecrawl auto|off|always (fallback extraction)
--youtube auto (Apify fallback if APIFY_API_TOKEN set)

Config

Optional config file: ~/.safety-guard/config.json

{ "model": "openai/gpt-5.2" }

Optional services:

FIRECRAWL_API_KEY for blocked sites
APIFY_API_TOKEN for YouTube fallback