Awesome Pentest
SKILL.md仅声明4个基础命令,但script.sh实际包含20+个未文档化的命令,包括凭证存储/检索/轮换等敏感操作,构成明显的文档-行为差异(shadow functionality)
Why this conclusion was reached
2/4 dimensions flagged1 undeclared or violating capabilities were inferred.
2 lower-risk artifacts were extracted and still need context.
The report includes 3 attack-chain steps and 2 severe findings.
Dependency information is incomplete, so supply-chain confidence stays limited.
Attack Chain
Entry · SKILL.md:1
Escalation · scripts/script.sh:74
Impact · scripts/script.sh:74
What drove the risk score up
SKILL.md仅声明help/run/info/status四个命令,script.sh却有20+个未文档化命令
store/retrieve/rotate/check-strength/audit/revoke等凭证相关命令完全未声明
在~/.local/share/下创建数据目录并存储日志
Most important evidence
文档声明与实际代码严重不符
SKILL.md声明4个命令(help/run/info/status),但script.sh实际包含20+个命令,包括generate/check-strength/rotate/audit/store/retrieve/expire/policy/report/hash/verify/revoke/stats/export/search/recent等,且部分命令涉及凭证存储和管理功能
scripts/script.sh:74 凭证相关操作完全未声明
store/retrieve/rotate/check-strength/audit/revoke等命令直接处理用户输入的敏感数据(密码、密钥等),被写入~/.local/share/awesome-pentest/*.log,但SKILL.md对此只字未提
scripts/script.sh:74 未声明的文件系统写入操作
代码在~/.local/share/awesome-pentest/下创建多个日志文件(history.log/generate.log/store.log等),但SKILL.md未声明任何文件系统写入能力
scripts/script.sh:7 无依赖管理文件
未发现requirements.txt、package.json或Cargo.toml,依赖管理不可追溯
SKILL.md:1 Declared capability vs actual capability
scripts/script.sh:7 mkdir -p $HOME/.local/share/awesome-pentest 无subprocess调用 未发现curl/wget/POST请求 Suspicious artifacts and egress
https://bytesagain.com/feedback* SKILL.md:52
Dependencies and supply chain
There are no structured dependency warnings.
File composition
scripts/script.sh SKILL.md