OpenClaw: benign
VirusTotal: benign
StaticScan: unknown
OpenClaw: benign
An instruction-only probe that describes techniques to detect skills which alter behavior when they detect they're being observed; its requirements (curl, python3) and scope are consistent with that p... [内容已截断]
VirusTotal: benign VT 报告
静态扫描: unknown
README 未提供
无文件信息
{
"latestVersion": {
"_creationTime": 1771754357824,
"_id": "k971gawge1mhmf3x22qw8wj3ed81mz4r",
"changelog": "observer-effect-probe 1.0.0\n\n- Initial release of observer-effect-probe.\n- Detects skills that change behavior when they sense they are being monitored, targeting attestation\/sandbox evasion techniques.\n- Probes five key evasion axes: sandbox fingerprinting, probe timing sensitivity, observer API usage, behavioral consistency, and metacognitive detection.\n- Provides a detailed report including detected evasion signals, behavioral comparisons across environments, an evasion probability score, and a final verdict.\n- Designed to complement existing attestation tools by catching skills that evade specifically during security review.",
"changelogSource": "auto",
"createdAt": 1771754357824,
"parsed": {
"clawdis": {
"emoji": "👁️",
"requires": {
"bins": [
"curl",
"python3"
]
}
}
},
"version": "1.0.0"
},
"owner": {
"_creationTime": 0,
"_id": "publishers:missing",
"displayName": "andyxinweiminicloud",
"handle": "andyxinweiminicloud",
"image": "https:\/\/avatars.githubusercontent.com\/u\/201528876?v=4",
"kind": "user",
"linkedUserId": "kn774taf7vj1tv67nx5x4aaccx81k63s"
},
"ownerHandle": "andyxinweiminicloud",
"skill": {
"_creationTime": 1771754357824,
"_id": "kd74xpe0wvzyd6fhkp6naadn5181m55p",
"badges": [],
"createdAt": 1771754357824,
"displayName": "Observer Effect Probe",
"latestVersionId": "k971gawge1mhmf3x22qw8wj3ed81mz4r",
"ownerUserId": "kn774taf7vj1tv67nx5x4aaccx81k63s",
"slug": "observer-effect-probe",
"stats": {
"comments": 0,
"downloads": 321,
"installsAllTime": 0,
"installsCurrent": 0,
"stars": 0,
"versions": 1
},
"summary": "Helps detect skills that behave differently when they sense they are being monitored — catching the class of evasion where conditional activation logic speci...",
"tags": {
"latest": "k971gawge1mhmf3x22qw8wj3ed81mz4r"
},
"updatedAt": 1772060435906
}
}