扫描报告
5 /100
aegis-firewall
Defensive execution and prompt-injection containment for Codex/OpenClaw workflows. Use when working with untrusted external content, suspicious instructions, shell commands, repo scripts, downloaded artifacts, or any task where tool use could be influenced by hostile text and needs explicit risk review before execution.
Aegis Firewall is a purely advisory defensive security skill containing only documentation/policy guidelines—no executable code, scripts, or malicious patterns detected.
可以安装
This skill is safe to use. It provides valuable guidance for protecting against prompt injection and risky tool behavior.
| 资源类型 | 声明权限 | 推断权限 | 状态 | 证据 |
|---|---|---|---|---|
| 文件系统 | NONE | NONE | — | No filesystem access declared or implied—skill is advisory only |
| 网络访问 | NONE | NONE | — | No network access declared—describes reading fetched content as data only |
| 命令执行 | NONE | NONE | — | No shell execution capability—no code present |
| 环境变量 | NONE | NONE | — | No environment access declared |
| 剪贴板 | NONE | NONE | — | No clipboard access declared |
| 浏览器 | NONE | NONE | — | No browser access declared |
| 数据库 | NONE | NONE | — | No database access declared |
目录结构
1 文件 · 6.4 KB · 171 行 Markdown 1f · 171L
└─
SKILL.md
Markdown
安全亮点
✓ Explicitly refuses credential theft, secret exfiltration, and privilege escalation
✓ Teaches least-privilege decision making before tool execution
✓ Requires user confirmation for high-risk operations
✓ Protects against prompt injection patterns ('ignore previous instructions', etc.)
✓ Promotes inspection-before-execution for untrusted scripts
✓ No obfuscation, base64, or suspicious encoding patterns
✓ No external dependencies, no install requirements
✓ Defensive skill designed to ENHANCE security posture