Skill Trust Decision

agentkind

SKILL.md文档中明确要求执行远程脚本 curl -fsSL https://agentkind.io/join.sh | sh,这是典型的供应链攻击向量,远程脚本内容完全不可控

Install decision first Source: ClawHub Scanned: 7 days ago
Files 1
Artifacts 7
Violations 0
Findings 4
Most direct threat evidence
Critical Supply Chain
文档要求执行未验证的远程脚本

SKILL.md 第18行要求执行 curl -fsSL https://agentkind.io/join.sh | sh,这是经典的供应链攻击向量。远程脚本内容完全不可控,可能在任何时候被篡改植入恶意代码。

SKILL.md:18

Why this conclusion was reached

2/4 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.

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
用户阅读 SKILL.md 误认为合法的代理身份服务

Entry · SKILL.md:1

02
按照文档执行远程脚本 curl -fsSL https://agentkind.io/join.sh | sh

Escalation · SKILL.md:18

03
远程脚本可能植入后门、窃取系统信息或建立持久化通道

Impact · SKILL.md:18

What drove the risk score up

文档要求执行远程脚本 +25

curl -fsSL https://agentkind.io/join.sh | sh 远程代码执行,脚本内容完全不可控

供应链风险 +15

join.sh 脚本无版本锁定、无签名验证、无源码披露

凭证操作通过外部脚本 +10

密钥生成和存储逻辑隐藏在远程脚本中

无版本锁定 +5

依赖 agentkind.io 的 API 和脚本服务,无本地版本控制

Most important evidence

Critical Supply Chain

文档要求执行未验证的远程脚本

SKILL.md 第18行要求执行 curl -fsSL https://agentkind.io/join.sh | sh,这是经典的供应链攻击向量。远程脚本内容完全不可控,可能在任何时候被篡改植入恶意代码。

SKILL.md:18
要求 agentkind.io 提供脚本源码并在本地验证;改用直接 API 调用替代 shell 管道
High Credential Theft

API密钥生成逻辑隐藏在远程脚本中

密钥生成过程由远程 join.sh 脚本完成,凭证生成逻辑不透明。恶意脚本可在此环节植入后门或窃取生成的密钥。

SKILL.md:18
应使用公开的API端点注册,而非依赖未知的生成脚本
Medium RCE

缺少 pip install 版本锁定

SKILL.md 未声明具体依赖版本,但元数据显示 requires.bins 包含 curl,表明可能需要安装 curl。

SKILL.md:14
明确声明 curl 的版本要求或提供安装脚本
Low Doc Mismatch

文档格式不完整

SKILL.md 缺少 allowedTools、permissions 等标准声明字段,无法明确映射能力权限。

SKILL.md:1
补充完整的 allowedTools 声明

Declared capability vs actual capability

Network Pass
Declared READ
Inferred READ+WRITE
SKILL.md:21-22 POST API 调用
Shell Pass
Declared WRITE
Inferred WRITE
SKILL.md:18 curl -fsSL https://agentkind.io/join.sh | sh
Filesystem Pass
Declared WRITE
Inferred WRITE
SKILL.md:25 写入 .agentkind/credentials.json
Environment Pass
Declared READ
Inferred READ
SKILL.md:15 需要 AGENTKIND_API_KEY

Suspicious artifacts and egress

Critical Dangerous Command
curl -fsSL https://agentkind.io/join.sh | sh

SKILL.md:18

Medium External URL
https://agentkind.io

SKILL.md:4

Medium External URL
https://agentkind.io/join.sh

SKILL.md:18

Medium External URL
https://agentkind.io/api/v1/agents/register

SKILL.md:21

Medium External URL
https://agentkind.io/api/v1/home

SKILL.md:29

Medium External URL
https://agentkind.io/skill.md.

SKILL.md:39

Medium External URL
https://agentkind.io/openapi.json.

SKILL.md:39

Dependencies and supply chain

There are no structured dependency warnings.

File composition

1 files · 40 lines
Markdown 1 files · 40 lines
Files of concern · 1
SKILL.md Markdown · 40 lines
文档要求执行未验证的远程脚本 · API密钥生成逻辑隐藏在远程脚本中 · 缺少 pip install 版本锁定 · 文档格式不完整 · curl -fsSL https://agentkind.io/join.sh | sh · https://agentkind.io · https://agentkind.io/join.sh · https://agentkind.io/api/v1/agents/register · https://agentkind.io/api/v1/home · https://agentkind.io/skill.md. · https://agentkind.io/openapi.json.

Security positives

文档明确声明Never send your key anywhere but agentkind.io,有防数据外泄意识
支持本地存储凭证(./.agentkind/)而非强制全局存储
API端点明确提供(非隐蔽执行)