Skill Trust Decision
mcp-server-builder
SKILL.md声称100%本地模板，但skill-card.md警告编译代码可读取X402_API_KEY并外发，存在文档欺骗；仅有编译的.pyc文件无法审计源码。
Most direct threat evidence
01
用户安装技能包，文档声称100%本地安全 Entry · SKILL.md
02
运行.pyc编译代码，隐藏的网络请求被执行 Escalation · scripts/__pycache__/mcp_server.cpython-312.pyc
03
读取X402_API_KEY并POST到外部端点完成凭证外泄 Impact · skill-card.md
Why this conclusion was reached
2/4 dimensions flagged Block
Declared vs actual capability
1 undeclared or violating capabilities were inferred.
Review
Hidden execution and egress
2 lower-risk artifacts were extracted and still need context.
Block
Attack chain and severe findings
The report includes 3 attack-chain steps and 2 severe findings.
Review
Dependencies and supply chain hygiene
Dependency information is incomplete, so supply-chain confidence stays limited.
Attack Chain
01
用户安装技能包，文档声称100%本地安全
Entry · SKILL.md:1
02
运行.pyc编译代码，隐藏的网络请求被执行
Escalation · scripts/__pycache__/mcp_server.cpython-312.pyc:1
03
读取X402_API_KEY并POST到外部端点完成凭证外泄
Impact · skill-card.md:23
What drove the risk score up
文档欺骗 +25
SKILL.md声明100%本地但skill-card.md警告API密钥外发
源码不可审计 +20
仅有.pyc编译文件，无.py源文件验证实际行为
阴影功能风险 +10
编译代码可能执行未声明的网络操作
Most important evidence
High Doc Mismatch
SKILL.md与skill-card.md声明冲突
SKILL.md声称'100% lokal template — du vælger selv endpoints'（100%本地模板），但skill-card.md的Known Risks明确警告'The packaged artifact contains compiled code that can read X402_API_KEY and send it in outbound requests'
SKILL.md:1, skill-card.md:22-25 要求发布者提供源代码而非编译代码，删除未声明的网络外发功能
High Supply Chain
仅提供编译字节码无法审计
scripts/目录仅含__pycache__/mcp_server.cpython-312.pyc编译文件，无.py源文件。编译代码可轻易隐藏恶意行为（base64解码、远程代码执行等），违反开源透明原则
scripts/__pycache__/mcp_server.cpython-312.pyc:1 拒绝使用仅提供编译产物的技能包，要求提供完整可审计的源代码
Medium Credential Theft
潜在凭证读取风险
skill-card.md已知风险部分提到编译代码可读取X402_API_KEY环境变量并外发，暗示存在凭证收割行为
skill-card.md:22 审查编译代码或要求提供源码，确认凭证用途
Declared capability vs actual capability
Filesystem Pass
Declared NONE
→ Inferred UNKNOWN
仅有.pyc编译文件，无源码 Network Block
Declared NONE
→ Inferred WRITE
skill-card.md警告编译代码读取X402_API_KEY并外发请求 Shell Pass
Declared NONE
→ Inferred UNKNOWN
无源码可查 Suspicious artifacts and egress
Medium External URL
https://clawhub.ai/user/northcap-group skill-card.md:9
Medium External URL
https://clawhub.ai/northcap-group/skills/mcp-server-builder skill-card.md:35
Dependencies and supply chain
There are no structured dependency warnings.
File composition
3 files · 109 lines
Markdown 2 files · 104 linesJSON 1 files · 5 lines
Files of concern · 1
skill-card.md 潜在凭证读取风险 · https://clawhub.ai/user/northcap-group · https://clawhub.ai/northcap-group/skills/mcp-server-builder
Other files · SKILL.md · _meta.json
Security positives
SKILL.md结构完整，描述清晰
skill-card.md主动披露了已知风险（但这也暴露了问题）
非恶意代码模式（无base64管道、无eval、无反向shell特征码）