Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Gen Design
v1.0.0通过交互式技术选型生成系统设计文档。基于领域知识库进行数据模型设计和集成设计，确保设计与 PRD 数据流对齐。仅当用户要「生成系统设计文档」时使用；分析现有代码库请用 analyze，从代码反推设计主线请用 analyze --phase=deep。
⭐ 0· 63·0 current·0 all-time
byLiu Feng@lf951515851
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill's name and description match the instructions: it interactively generates system design docs using a domain knowledgebase. However, the skill repeatedly expects to load files from skills/libs/domain-knowledge/{domain}.md and architecture-options.md that are not included in the published bundle. Relying on external knowledgebase files outside the skill package is plausible for this use case but is an important dependency mismatch (the skill declares no required config paths yet references them).
Instruction Scope
SKILL.md only instructs reading PRD inputs and knowledgebase files, performing interactive Q&A, auto-filling design sections, and optionally calling a /validate step. It does not ask for credentials or network uploads, nor does it instruct reading unrelated system artifacts like shell history. Concern: the skill explicitly loads files from a path (skills/libs/domain-knowledge/...) that lies outside the skill's own file list — this means it will attempt to read platform-level files, which expands its effective scope. It also instructs 'auto-continue' behavior (call /validate when user replies yes or does not say pause), which enables chaining to other skills.
Install Mechanism
No install spec and no code files that execute on install — instruction-only. This is the lowest install risk: nothing is downloaded or written by an installer. The files provided are templates and prompts only.
Credentials
The skill declares no required environment variables, no credentials, and no config paths. That is proportionate to a design-document generation skill. The only potential gap is that the instructions expect to load domain-knowledge files from platform paths that were not declared as required; this is a benign capability for the stated purpose but should be verified to ensure no unexpected file access.
Persistence & Privilege
always is false and disable-model-invocation is false (normal). The skill's instructions include an 'automatic continue' behavior that, if supported by the platform, will autonomously invoke another skill (/validate). Autonomous chaining is expected for skills, but you should be aware this grants the skill the ability to trigger additional actions when the user responds 'yes' or does not explicitly pause.
What to consider before installing
This skill looks like a legitimate design-doc generator, but before installing: 1) Ask where the 'domain-knowledge' files live on your platform — the package does not include them but the skill will try to load skills/libs/domain-knowledge/*.md. Confirm those files are trustworthy and do not contain sensitive data. 2) Confirm the platform enforces file-access boundaries so the skill cannot read unrelated files. 3) Be aware of the 'auto-continue' behavior that will call /validate (another skill) if the user replies 'yes' or doesn't explicitly pause; if you want to restrict automatic chaining, disable or require explicit consent. 4) Because this is instruction-only, there's no installer risk, but verify the knowledgebase and any target repo paths (frontendRoot/backendRoot) the skill will create/write to before allowing it to modify your workspace. If you want more certainty, request the author include or document the required knowledgebase files and explicit file-access requirements.Like a lobster shell, security has layers — review code before you run it.
latestvk97fzy8qctt2fj98a9ahsm92m983t3j3
License
MIT-0
Free to use, modify, and redistribute. No attribution required.