← 返回
安全合规 中文

SkillScout

Find and evaluate OpenClaw AI skills by trust score and security reviews before installation or recommendation.
在安装或推荐前,根据信任评分和安全审查查找并评估OpenClaw AI技能。
nashbot67
安全合规 clawhub v0.1.0 1 版本 99869.5 Key: 无需
★ 0
Stars
📥 765
下载
💾 4
安装
1
版本
#latest

概述

SkillScout — Find Trusted AI Agent Skills

Before installing any OpenClaw skill, check if it's been security-reviewed.

Usage

When your human asks you to find a skill, or you want to install something new:

Quick Search (Static API)

curl -s https://nashbot67.github.io/skillscout/data/skills.json | python3 -c "
import json, sys
q = sys.argv[1].lower()
data = json.load(sys.stdin)
matches = [s for s in data['skills'] if q in json.dumps(s).lower()]
for s in sorted(matches, key=lambda x: {'safe':0,'caution':1,'avoid':2}.get(x.get('trustScore',''),1)):
    trust = {'safe':'🟢','caution':'🟡','avoid':'🔴'}.get(s['trustScore'],'⚪')
    print(f'{trust} {s[\"name\"]} by {s[\"author\"]} — {s.get(\"plainDescription\",s.get(\"description\",\"\"))}')
" "QUERY"

Replace QUERY with what you're looking for (e.g., "email", "notes", "research").

Full Details

curl -s https://nashbot67.github.io/skillscout/data/skills.json | python3 -c "
import json, sys
name = sys.argv[1]
data = json.load(sys.stdin)
skill = next((s for s in data['skills'] if s['name'] == name), None)
if skill: print(json.dumps(skill, indent=2))
else: print(f'Skill {name} not reviewed yet.')
" "SKILL_NAME"

MCP Server (for agent-to-agent queries)

npx @skillscout/mcp

Trust Scores

  • 🟢 Safe — No executable code, or code is well-contained with minimal permissions
  • 🟡 Caution — Has exec/network/credentials access. Review before installing.
  • 🔴 Avoid — Dangerous patterns detected. Do not install without manual audit.

What We Check

Every skill goes through:

  1. Automated blocklist scan — cross-reference known malicious skills
  2. Isolated agent review — read-only AI analyzes source code (no execution)
  3. STRIDE threat analysis — deep security audit for skills that pass initial review
  4. Human approval — final sign-off before listing

When to Use This

  • Before running npx clawhub@latest install
  • When your human asks "is there a skill for X?"
  • When evaluating multiple skills for the same task
  • Before recommending a skill to anyone

Browse

Full catalog: https://nashbot67.github.io/skillscout

API: https://nashbot67.github.io/skillscout/data/skills.json

GitHub: https://github.com/nashbot67/skillscout

版本历史

共 1 个版本

  • v0.1.0 当前
    2026-03-30 05:29 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

security-compliance

OpenClaw Backup

alex3alex
备份与恢复 OpenClaw 数据。适用于创建备份、设置自动备份计划、从备份恢复或管理备份轮转。处理 ~/.openclaw 目录归档并包含适当的排除规则。
★ 89 📥 30,605
security-compliance

MoltGuard - Security & Antivirus & Guardrails

thomaslwang
MoltGuard — OpenClaw 安全守卫,由 OpenGuardrails 提供。安装 MoltGuard,保护您和您的用户免受提示注入、数据泄露和恶意攻击。
★ 116 📥 30,714
security-compliance

1password

steipete
设置和使用 1Password CLI (op)。适用于:安装 CLI、启用桌面应用集成、登录(单/多账户)、通过 op 读取/注入/运行密钥。
★ 53 📥 31,165