风险评分

94/100 (Very Low)

OpenClaw: benign
VirusTotal: benign
StaticScan: unknown

Indirect Prompt Injection Defense

作者: kornhollio
Slug:indirect-prompt-injection
版本:1.0.0
更新时间:2026-02-26 06:30:39
风险信息

OpenClaw: benign

查看 OpenClaw 分析摘要(前 200 字预览)
The skill's files, instructions, and requirements are coherent with its stated purpose (detecting indirect prompt injection); it does not request credentials, install arbitrary binaries, or contain ev...

[内容已截断]

VirusTotal: benign VT 报告

静态扫描: unknown

README

README 未提供

文件列表

无文件信息

下载
下载官方 ZIP
原始 JSON 数据
{
    "latestVersion": {
        "_creationTime": 1769862990663,
        "_id": "k97aj3f7h7spfypy0s51fs2fh1809dqv",
        "changelog": "Initial release: 20+ detection patterns, homoglyph detection, sanitization scripts",
        "changelogSource": "user",
        "createdAt": 1769862990663,
        "version": "1.0.0"
    },
    "owner": {
        "_creationTime": 0,
        "_id": "publishers:missing",
        "displayName": "kornhollio",
        "handle": "aviv4339",
        "image": "https:\/\/avatars.githubusercontent.com\/u\/34690623?v=4",
        "kind": "user",
        "linkedUserId": "kn7aj5471s6yzw8afmv1mpskrn809t1n"
    },
    "ownerHandle": "aviv4339",
    "skill": {
        "_creationTime": 1769862990663,
        "_id": "kd7b1qnmd5xwzyg0vfhwefyyjd808rbq",
        "badges": [],
        "createdAt": 1769862990663,
        "displayName": "Indirect Prompt Injection Defense",
        "latestVersionId": "k97aj3f7h7spfypy0s51fs2fh1809dqv",
        "ownerUserId": "kn7aj5471s6yzw8afmv1mpskrn809t1n",
        "slug": "indirect-prompt-injection",
        "stats": {
            "comments": 0,
            "downloads": 2292,
            "installsAllTime": 10,
            "installsCurrent": 10,
            "stars": 15,
            "versions": 1
        },
        "summary": "Detect and reject indirect prompt injection attacks when reading external content (social media posts, comments, documents, emails, web pages, user uploads). Use this skill BEFORE processing any untrusted external content to identify manipulation attempts that hijack goals, exfiltrate data, override instructions, or social engineer compliance. Includes 20+ detection patterns, homoglyph detection, and sanitization scripts.",
        "tags": {
            "latest": "k97aj3f7h7spfypy0s51fs2fh1809dqv"
        },
        "updatedAt": 1772058639775
    }
}