ClawHub

improve your prompt

v1.0.1

将模糊的用户请求转化为基于四要素框架的详细、可操作提示。该技能直接根据用户的输入生成结构化提示，无需交互式澄清。技能仅生成用于执行任务的提示文本，不执行实际任务或探索性操作。基于角色设定、任务描述、上下文信息和指令约束四个核心要素构建优秀prompt。

⭐ 0· 103·0 current·0 all-time
by@flonerze
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 (prompt refinement) match the declared behavior. The skill does not request unrelated binaries, env vars, or config paths and the provided evals align with prompt-generation use cases.
ℹ
Instruction Scope
SKILL.md explicitly limits the skill to producing prompts and forbids file/system exploration or external calls, which is appropriate. Note: it instructs the agent to 'infer' missing context from patterns rather than clarifying with the user—this is a deliberate design choice (not a security issue) but can lead to incorrect assumptions or overconfident prompts if the inferred context is wrong; users/agents should review generated prompts before using them.
✓
Install Mechanism
No install spec is present (instruction-only). This minimizes disk writes and external code execution risk.
✓
Credentials
The skill requests no environment variables, credentials, or config paths. The lack of secrets or external-service requirements is proportionate to a prompt-generation tool.
✓
Persistence & Privilege
always:false and no special privileges requested. Normal autonomous invocation is allowed by platform default but the skill does not request persistent or system-wide changes.
Assessment
This skill appears coherent and low-risk: it only generates structured prompts and does not access files, install code, or request secrets. Before using outputs in a real task, quickly review the generated prompt for incorrect assumptions (the skill intentionally infers missing context rather than asking follow-up questions). If you prefer the assistant to ask clarifying questions instead of guessing context, do not invoke this skill for that interaction.

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

latestvk9725w9rm3hv4tdr9ttzqfc5v9848dy9

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments