Skill Trust Decision

plori

纯文档型技能,通过shell管道安装CLI存在供应链风险,但skill-card已声明警告,且skill本身无代码执行能力

Install decision first Source: ClawHub Scanned: 9 days ago
Files 2
Artifacts 17
Violations 0
Findings 2
Most direct threat evidence
Critical Dangerous Command
curl -fsSL https://plori.ai/install.sh | sh

Why this conclusion was reached

1/4 dimensions flagged
Pass
Declared vs actual capability

Declared resources and inferred behavior are broadly aligned.

Block
Hidden execution and egress

1 high-risk artifacts or egress signals were extracted.

Pass
Attack chain and severe findings

There is no explicit malicious chain in the report.

Review
Dependencies and supply chain hygiene

Dependency information is incomplete, so supply-chain confidence stays limited.

What drove the risk score up

shell管道安装命令 +25

curl -fsSL https://plori.ai/install.sh | sh 存在供应链攻击风险

无脚本代码审查 +0

技能仅含文档,无可执行代码

风险已声明 +-10

skill-card.md明确标注了管道安装风险并建议MCP替代

Most important evidence

Medium Supply Chain

Shell管道安装命令存在供应链风险

SKILL.md:131 包含 curl -fsSL https://plori.ai/install.sh | sh 命令,将远程脚本直接管道执行Shell是典型供应链攻击向量。skill-card.md已声明此风险并建议使用MCP方式连接。

SKILL.md:131
优先使用MCP连接方式;如需CLI安装,应先curl下载脚本审查内容后再执行
Low Priv Escalation

Skill功能与声明权限可能不符

skill-card未声明allowedTools,但SKILL.md描述的OAuth流程需要向外部端点发送敏感数据(device_code),且MCP工具可操作工作区文件。

SKILL.md:52
在skill-card.md中明确声明network:WRITE权限和OAuth交互行为

Declared capability vs actual capability

Network Pass
Declared READ
→
Inferred READ
SKILL.md:52-55 MCP API调用
Shell Pass
Declared NONE
→
Inferred NONE
无shell脚本代码
Filesystem Pass
Declared NONE
→
Inferred READ
SKILL.md:81 上传文件工具声明

Suspicious artifacts and egress

Critical Dangerous Command
curl -fsSL https://plori.ai/install.sh | sh

SKILL.md:131

Medium External URL
https://api.plori.ai/mcp

SKILL.md:20

Medium External URL
https://api.plori.ai/.well-known/oauth-protected-resource

SKILL.md:23

Medium External URL
https://plori.ai.

SKILL.md:25

Medium External URL
https://api.plori.ai/oauth/pair

SKILL.md:57

Medium External URL
https://api.plori.ai/oauth/pair/poll

SKILL.md:65

Medium External URL
https://plori.ai/install.sh

SKILL.md:131

Medium External URL
https://plori.ai/install.ps1

SKILL.md:131

Medium External URL
https://api.plori.ai/v1

SKILL.md:138

Medium External URL
https://plori.ai/auth.md

SKILL.md:138

Medium External URL
https://plori.ai/docs/cli

SKILL.md:282

Medium External URL
https://plori.ai/agents.md

SKILL.md:322

Dependencies and supply chain

There are no structured dependency warnings.

File composition

2 files · 382 lines
Markdown 2 files · 382 lines
Files of concern · 2
SKILL.md Markdown · 326 lines
Shell管道安装命令存在供应链风险 · Skill功能与声明权限可能不符 · curl -fsSL https://plori.ai/install.sh | sh · https://api.plori.ai/mcp · https://api.plori.ai/.well-known/oauth-protected-resource · https://plori.ai. · https://api.plori.ai/oauth/pair · https://api.plori.ai/oauth/pair/poll · https://plori.ai/install.sh · https://plori.ai/install.ps1 · https://api.plori.ai/v1 · https://plori.ai/auth.md · https://plori.ai/docs/cli · https://plori.ai/agents.md · https://plori.ai/mcp · https://www.npmjs.com/package/@plori/cli · https://plori.ai/llms.txt
skill-card.md Markdown · 56 lines
https://clawhub.ai/user/yuchen-plori · https://clawhub.ai/yuchen-plori/skills/plori

Security positives

skill-card.md明确声明了shell管道安装的供应链风险
文档推荐使用MCP方式连接,避免直接shell安装
无本地可执行代码,无法在目标系统植入后门
OAuth流程使用标准PKCE,凭证不经过AI Agent直接传递