雀漫
v1.0.1将长篇小说转换为 Seedance 2.0 平台的 15 秒动态漫画分镜脚本。 包含剧情容量分析、旁白与台词严格区分、角色卡(纯文本)、6-8 个分镜头、 微动作设计、镜头运动、配音音效。专为 AI 视频生成优化,无 markdown 格式符号。
⭐ 0· 130·0 current·0 all-time
byfslong@fslong520
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (convert long-form fiction into 15s Seedance 2.0 dynamic-comic scripts) matches the contents of SKILL.md and the supporting docs. No unexpected binaries, env vars, or config paths are requested.
Instruction Scope
All runtime instructions are text-based and self-contained: analyze input text, recommend art/style/canvas/audio, and output a plaintext storyboard. No network endpoints, credentials, or file I/O are required. Minor internal inconsistencies exist: e.g., Technical specifications and Checklist state output must be '纯文本' and '禁止使用任何 markdown 符号', but the checklist also mandates wrapping the script in ```plaintext code fences (a markdown construct); Stage 8 suggests '保存分镜脚本到本地' while the checklist also states '本技能不读取或写入任何文件'. These are functional/instructional contradictions (not evidence of exfiltration) that could confuse automated agents.
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest-risk install posture. There are no download URLs, packages, or extracted archives.
Credentials
The skill requests no environment variables, secrets, or config paths. Nothing asks for unrelated credentials or broad access.
Persistence & Privilege
always is false and the skill does not request persistent system presence or modify other skills. Autonomous invocation is allowed by default (disable-model-invocation is false) — this is platform default and not combined with other red flags.
Assessment
This skill appears coherent and low-risk: it only instructs the model how to convert provided text into a Seedance-style 15s storyboard and requests no credentials or installs. Before installing or using it: 1) test with non-sensitive/sample text to confirm the agent follows the 'plaintext' output rules (note the docs contradict themselves about triple-backtick wrapping vs forbidding markdown); 2) review outputs before copying them into any external Seedance account or saving locally (Stage 8 mentions saving locally even though the checklist says no file I/O); 3) avoid submitting copyrighted or personally identifiable material unless you have the right to do so; and 4) if you plan to let the agent invoke this skill autonomously, be aware that it may generate and submit storyboards into conversation flows — review policies for any automated publication step. If you want stronger assurance, request a clarified SKILL.md that resolves the markdown / file I/O contradictions and explicitly states whether any automatic network integration with Seedance exists.Like a lobster shell, security has layers — review code before you run it.
latestvk9798wq7d74v066wwd2kmv1yrx83m219
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
