Design
v1.2.1生成定制化的设计项目 Skill（含 SKILL.md + references/），为用户的设计过程提供从需求定义到 交付的持续引导。触发场景：用户表达想做设计相关工作（"我想设计 X""做个 X 的方案" "帮我设计一个 X""出个设计""改版 X""重新做一下 X""想想 X 怎么做""design X"...
⭐ 1· 140·0 current·0 all-time
byBalancegogogo@balancegsr
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The skill claims to generate a design-project Skill and guide a design process; its runtime instructions (collecting user requirements, templating SKILL.md, creating references/, reading/writing project files) match that purpose. The directory probing to detect an appropriate skill install prefix (workspace and home skill paths) is aligned with the stated 'install into workspace' behavior.
Instruction Scope
Instructions direct the agent to read and load project files (plan.md, decisions.md and other produced artifacts) and to scan for existing skill directories under the workspace and user home. This is coherent for a file-backed memory design assistant, but it means the agent will read arbitrary files under the project directory (and will enumerate certain paths in the user home). That has privacy implications if sensitive files are present in those locations.
Install Mechanism
No install spec and no code files — the skill is instruction-only. Nothing will be downloaded or executed from external URLs. This is the lowest-risk install model.
Credentials
The skill requests no environment variables, credentials, or external secrets. Its need for filesystem access (read/write/create under detected skill prefix and project_dir) is proportionate to its purpose of creating and maintaining project files.
Persistence & Privilege
The skill will persistently write files and directories into the workspace (and may write under a detected global skills prefix in the user's home). It does not request 'always: true' nor claim system-wide privilege, and it indicates it will create its own skill folder rather than modify other skills. Still, it is persistent (writes on disk) so users should expect permanent artifacts.
Assessment
This skill appears coherent and does what it says: it will collect brief requirements, generate a templated SKILL.md and supporting references, and create a project folder structure in your workspace (or package as ZIP). Before installing or allowing the agent to run in your real workspace: 1) Review the generated SKILL.md preview that the Creator shows before it writes files. 2) Run it in an isolated or dedicated project folder (not your home or an existing repo with secrets). The skill will scan common skill-paths under your workspace and home to pick an install prefix — if you do not want it to probe your home, run it in a clean workspace or choose ZIP packaging and install manually. 3) Be aware the agent will read and load files from the project_dir (plan.md, decisions.md and other outputs) into its context; remove or avoid placing secrets in those files. 4) Prefer ZIP delivery if you want to inspect files before they are written into any global skill path. If you are concerned about filesystem access, restrict the agent's file write/read permissions or run in a disposable environment.Like a lobster shell, security has layers — review code before you run it.
latestvk97csjymmndetvecm6cdaje1xs83w35w
License
MIT-0
Free to use, modify, and redistribute. No attribution required.