知识卡片提示词生成器

v1.0.0

将简短文本转换为知识卡片图片的提示词生成器(社区精简版)。基于严格的格式规则,自动分析输入内容并生成符合视觉笔记标准的图片提示词。

0· 58·0 current·0 all-time
by八点虾@spiderzhcl
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description align with the files and runtime instructions: the SKILL.md, README and reference docs all describe generating visual-note image prompts and enforcing a strict format. There are no unrelated env vars, binaries, or installs required.
Instruction Scope
The instructions stay within the stated purpose (generate formatted prompts). However, the skill enforces content-transforming rules that may surprise users: it requires every output field order and formatting, forces wrapping user text in backticks, mandates adding the author signature "By 八点虾" to every generated image, and forbids percentage symbols/numeric coordinates. There are internal inconsistencies (e.g., '文字内容必须与用户输入保持一致' vs. strict forbidding of percentage numbers in [Negative Prompt]) that can cause the skill to alter or reject parts of user input (numbers/percent-style content). This is a functional/semantic concern (not a network/secret exfiltration issue).
Install Mechanism
The skill is instruction-only and has no install spec (lowest attack surface). The README includes an optional git clone URL (https://github.com/spiderzhcl/visual-note-prompt) and a manual installation path, which would download code from an external source if followed — that is outside the platform's install spec and not required to run the instruction-only skill, but it's an actionable external link users may choose to follow.
Credentials
No environment variables, credentials, or config paths are requested. The skill does not ask for unrelated secrets or permissions.
Persistence & Privilege
Flags show default behavior (not always: true). The skill is user-invocable and allowed to be called autonomously by the model (platform default). It does not request elevated or persistent privileges.
Assessment
This skill appears coherent and low-risk from a system-security perspective because it is instruction-only and requests no credentials. Before installing/using it, consider: (1) it will add a mandatory author signature "By 八点虾" to every generated prompt/image — if you don't want that attribution, this skill will not respect that preference; (2) it enforces strict formatting and may modify or reject numeric/percentage content (there are conflicting rules about keeping user text unchanged vs. banning percent/numeric coordinates), so test with your real inputs (especially those containing numbers like "25分钟") to confirm behavior; (3) the README suggests cloning a GitHub repo if you want the files locally — downloading code from that external URL is optional but carries the usual risks of fetching code from the internet; (4) if you need different attribution, broader styles, or preservation of precise numeric content, look for or request a variant that does not enforce mandatory author tags or negative-prompt numeric bans.

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

latestvk97bwssnc32rf0w6enkpym7p3184cdhq

License

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

Comments