Core Brainstorming

v0.1.0

在任何创造性工作之前必须使用 - 创建功能、构建组件、添加新功能或修改行为。通过协作对话探索用户意图、需求和设计。

0· 458·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description match the instructions: the skill is meant to run a collaborative design brainstorm and the SKILL.md's steps (gather context, ask questions, propose options, write design) align with that purpose.
Instruction Scope
The instructions tell the agent to inspect the current project state (files, docs, recent commits) and to write a design file under docs/plans/... and commit it to git. Those filesystem and git actions are expected for an in-repo brainstorming workflow, but they grant the skill read/write access to the repository and history—ensure you run it only in trusted projects.
Install Mechanism
No install spec or code is provided (instruction-only), so nothing will be downloaded or written by an installer.
Credentials
The skill declares no environment variables or credentials. However, committing to git and reading recent commits implies the agent will need repository access and possibly configured git credentials; this is plausible but not explicitly declared.
Persistence & Privilege
always is false and the skill does not request persistent privileges or to modify other skills. It does reference required/recommended sub-skills, which is normal.
Assessment
This skill is instruction-only and appears coherent for in-repo brainstorming. Before installing or invoking it: (1) only run it in repositories you trust because it will read files and recent commits and will create and git-commit design docs; (2) ensure the agent's runtime has the intended git configuration/credentials if you want commits to succeed; (3) verify you are comfortable with the skill invoking the declared sub-skill(s) (core-writing-plans and any recommended embedded/simulation helpers) since those may perform further actions; (4) if you need stricter limits, run the skill in a sandboxed workspace or temporarily revoke write access to the repository.

Like a lobster shell, security has layers — review code before you run it.

latestvk97c4xmn53a9zf7fcpv4n2rq5d82safw

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments