story-bozo
v1.0.0网文章节分镜转化技能。将小说章节内容或分镜内容转化为富有画面感、节奏感和悬念感的立体分镜。 使用场景： - 用户说"帮我优化这章的分镜"、"把这个章节改成分镜脚本" - 用户说"这个情节太平淡了，帮我增加爽感" - 用户说"根据这些分镜内容写成立体分镜" - 用户提供小说章节，要求增强画面感和节奏感 - 用户说"...
⭐ 0· 116·0 current·0 all-time
by@bozoyan
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (storyboard / '爽感' writing) line up with the contents: all files are writing guidance and examples. There are no unexpected environment variables, binaries, or external services requested that would be unrelated to a text-transformation writing skill.
Instruction Scope
SKILL.md provides concrete narrative transformation rules, templates, examples, and checklists. It does not tell the agent to read system files, access environment variables, or send data to external endpoints beyond the agent's normal model calls. No vague 'gather arbitrary context' instructions appear.
Install Mechanism
There is no install spec and no code files; the skill is instruction-only, so nothing is downloaded or written to disk by the skill itself.
Credentials
requires.env is empty and no credentials/config paths are requested. The skill's functionality (text-to-storyboard guidance) does not justify any secrets or privileged access.
Persistence & Privilege
always is false and the skill does not request elevated or persistent system privileges. Nothing in the files indicates modification of other skills or system-wide settings.
Assessment
This skill is internally consistent and low-risk because it is instruction-only writing guidance. Before installing, consider: (1) privacy: any manuscript or unpublished text you submit to the agent could be sent to the model—avoid submitting sensitive/confidential content; (2) copyright: the skill can rewrite user-provided copyrighted material—ensure you have rights to use that text; (3) content safety: the skill encourages creating high-impact '爽感' scenes which can include violent or sexual elements—validate outputs for policy/compliance before publishing; (4) quality control: the skill provides rules and templates but the final output quality depends on the model—test with representative examples to confirm tone and behavior; and (5) platform behavior: although the skill doesn't call external endpoints or require secrets, the agent platform will still send prompts to the model service per normal operation—if you need stricter data residency or auditability, check the platform's data handling policies. Overall this appears coherent and appropriate for its stated purpose.Like a lobster shell, security has layers — review code before you run it.
latestvk9751rw7z7v1wvgc37gf92ffqd839kfe
License
MIT-0
Free to use, modify, and redistribute. No attribution required.