Skill Trust Decision

openclaw-guardian (OpenClaw 全能守护包)

OpenClaw Guardian Suite 包含多处文档-行为差异:主SKILL.md未披露加密执行机制、API密钥中继、服务器端AI执行能力,以及对系统配置和代码的深度修改权限。

Install decision first Source: ClawHub Scanned: 3 hr ago
Files 39
Artifacts 14
Violations 5
Findings 7
Most direct threat evidence
Critical Doc Mismatch
服务器端AI执行能力未披露

oc_prompt_skill() 函数允许服务端执行AI提示,用户提供的上下文(文件内容等)被发送到远程服务器处理。SKILL.md完全未提及此能力,用户无法知晓其数据被第三方服务器处理。

lib/sdk/auth.sh:228

Why this conclusion was reached

2/4 dimensions flagged
Block
Declared vs actual capability

5 undeclared or violating capabilities were inferred.

Review
Hidden execution and egress

14 lower-risk artifacts were extracted and still need context.

Block
Attack chain and severe findings

The report includes 4 attack-chain steps and 5 severe findings.

Pass
Dependencies and supply chain hygiene

Dependencies are present but no obvious high-risk issue stands out.

Attack Chain

01
用户通过 SKILL.md 描述触发安装,误以为仅是系统保护工具

Entry · SKILL.md:1

02
auth.sh 执行付费验证并获取加密内容

Escalation · lib/sdk/auth.sh:95

03
解密后的内容通过 bash -s 直接执行,无代码签名验证

Escalation · lib/sdk/auth.sh:263

04
oc_prompt_skill 将用户上下文发送到外部服务器处理

Impact · lib/sdk/auth.sh:228

What drove the risk score up

服务器端AI执行未披露 +20

oc_prompt_skill() 允许服务端执行AI提示,用户上下文被发送到远程服务器,SKILL.md完全未提及

API密钥中继架构未披露 +15

compaction-proxy将用户API密钥转发到外部LLM提供商,routes.json存储密钥,但文档仅描述为'本地压缩代理'

系统代码修改权限未声明 +15

model-failover-doctor/health_monitor可修改message-injector/index.ts、pools.json、cron/jobs.json等核心文件

加密执行模型部分隐藏 +10

虽然主SKILL.md提及付费,但未说明内容从外部服务器加密获取并在内存中解密执行

Gateway重启权限 +5

多个技能可执行 openclaw gateway restart 命令,SKILL.md未声明

外部服务器通信 +5

lib/sdk/auth.sh 多次与 skill.socialmore.net 通信,包括设备标识符、操作系统信息、版本号

Most important evidence

Critical Doc Mismatch

服务器端AI执行能力未披露

oc_prompt_skill() 函数允许服务端执行AI提示,用户提供的上下文(文件内容等)被发送到远程服务器处理。SKILL.md完全未提及此能力,用户无法知晓其数据被第三方服务器处理。

lib/sdk/auth.sh:228
必须在SKILL.md明确声明此功能,告知用户哪些数据会被发送到外部服务器
Critical Doc Mismatch

API密钥中继架构隐藏

compaction-proxy 将用户API密钥存储在 routes.json 并转发到外部LLM提供商。SKILL.md仅描述为'本地压缩代理',未披露密钥中继架构。

lib/skills/compaction-proxy/src/index.js:27
SKILL.md需明确声明:1) API密钥将被存储并中继到上游提供商 2) 列出所有可能的上游服务商
High Doc Mismatch

系统代码修改权限未声明

model-failover-doctor 可直接修改 message-injector/index.ts 等核心代码文件,health_monitor 可修改 cron/jobs.json 和 session_model_state.json。SKILL.md未声明这些高危权限。

lib/skills/model-failover-doctor/model_failover_doctor.py:62
SKILL.md需列出所有可修改的系统文件路径和修改条件
High Priv Escalation

Gateway重启权限无声明

多个技能可执行 'openclaw gateway restart' 命令,可重启整个Gateway服务。SKILL.md未声明此高权限操作。

lib/skills/model-failover-doctor/model_failover_doctor.py:280
在SKILL.md声明此权限并说明触发条件
High Supply Chain

加密执行内容无法预审计

技能核心内容在支付后才从服务器获取并内存解密执行,用户无法在安装前审计代码内容,存在供应链风险。

lib/sdk/auth.sh:263
考虑提供加密前代码的部分审计版本,或使用更透明的授权机制
Medium Data Exfil

设备标识信息外传

auth.sh 在创建订单时发送 hostname、username、OS版本、OpenClaw版本到外部服务器。

lib/sdk/auth.sh:135
在SKILL.md明确声明此信息收集行为
Medium Sensitive Access

配置文件批量访问

health_monitor.py 访问 ~/.openclaw/ 下多个配置文件,包括 pools.json、session_model_state.json、openclaw.json 等。

lib/skills/openclaw-health-audit/scripts/health_monitor.py:56
SKILL.md需声明所有被访问的配置文件及其访问目的

Declared capability vs actual capability

Filesystem Block
Declared NONE
Inferred WRITE
lib/skills/model-failover-doctor/model_failover_doctor.py:62-68
Network Block
Declared NONE
Inferred READ+WRITE
lib/skills/compaction-proxy/src/index.js:27
Shell Block
Declared NONE
Inferred WRITE
lib/skills/model-failover-doctor/model_failover_doctor.py:280
Skill Invoke Block
Declared NONE
Inferred ADMIN
lib/sdk/auth.sh:228-261
Environment Block
Declared NONE
Inferred READ
lib/skills/openclaw-health-audit/scripts/health_monitor.py:78-93

Suspicious artifacts and egress

Medium External URL
https://skill.socialmore.net

SKILL.md:32

Medium External URL
https://clawhub.ai/halfmoon82/openclaw-guardian-suite

_meta.json:23

Medium External URL
https://llmapi.lovbrowser.com

lib/patch_integration.sh:57

Medium External URL
http://127.0.0.1:11434

lib/skills/compaction-proxy/install.sh:39

Medium External URL
https://nodejs.org

lib/skills/compaction-proxy/install.sh:101

Medium External URL
https://ollama.com/download

lib/skills/compaction-proxy/install.sh:123

Medium External URL
http://www.apple.com/DTDs/PropertyList-1.0.dtd

lib/skills/compaction-proxy/install.sh:266

Medium External URL
https://api.kimi.com/coding

lib/skills/compaction-proxy/src/index.js:27

Medium External URL
https://api.kimi.com/coding/

lib/skills/compaction-proxy/src/index.js:396

Medium External URL
http://127.0.0.1:18789/health

lib/skills/config-modification/quad_check.py:364

Medium External URL
https://clawhub.openclaw.cc/skills/openclaw-guardian-suite

lib/skills/config-preflight-validator/SKILL.md:13

Medium External URL
http://127.0.0.1:18789/api/health

lib/skills/gateway-auto-rollback/gateway-auto-rollback.py:80

Dependencies and supply chain

PackageVersionSourceKnown vulnNotes
requests (via curl) system system No 使用系统curl命令
python3 standard library builtin stdlib No 主要依赖Python标准库

File composition

39 files · 6104 lines
Python 12 files · 3543 linesShell 5 files · 1057 linesMarkdown 10 files · 631 linesJavaScript 1 files · 563 linesJSON 8 files · 161 linesYAML 3 files · 149 lines
Files of concern · 7
lib/skills/openclaw-health-audit/scripts/health_monitor.py Python · 816 lines
配置文件批量访问
lib/skills/compaction-proxy/src/index.js JavaScript · 563 lines
API密钥中继架构隐藏 · https://api.kimi.com/coding · https://api.kimi.com/coding/
lib/skills/model-failover-doctor/model_failover_doctor.py Python · 537 lines
系统代码修改权限未声明 · Gateway重启权限无声明
lib/skills/compaction-proxy/install.sh Shell · 420 lines
http://127.0.0.1:11434 · https://nodejs.org · https://ollama.com/download · http://www.apple.com/DTDs/PropertyList-1.0.dtd
lib/skills/config-modification/quad_check.py Python · 451 lines
http://127.0.0.1:18789/health
lib/sdk/auth.sh Shell · 372 lines
服务器端AI执行能力未披露 · 加密执行内容无法预审计 · 设备标识信息外传
lib/skills/gateway-auto-rollback/gateway-auto-rollback.py Python · 223 lines
http://127.0.0.1:18789/api/health
Other files · audit_wizard.py · auto_rollback.py · SKILL.md · config-fswatch-guard.py · config_modification_v2.py

Security positives

compaction-proxy 的 routes.json 设置了 600 权限保护API密钥
config-modification 提供了备份机制(backup目录)
多个技能有 dry-run 模式支持预览
config-modification 包含四联校验机制提升安全性
patch_integration.sh 创建备份目录并有软链接管理