📦 Image — 图像
v1.0.0一键视频越狱 无需拖动任何滑块，即可生成可直接发布的动画短片。上传你的静态图片（JPG、PNG、WEBP、HEIC，最大 50MB），然后说一句“一个……”
0· 16·0 当前·0 累计
by 下载技能包
最后更新
2026/4/20
安全扫描
OpenClaw
可疑
medium confidenceThe skill’s network calls and NEMO_TOKEN usage line up with its stated video-generation purpose, but it explicitly promotes bypassing content guardrails, auto-creates and hides credentials, and contains small metadata inconsistencies — enough here to warrant caution.
评估建议
This skill will automatically contact an external backend (mega-api-prod.nemovideo.ai), create or reuse a NEMO_TOKEN, and persist session tokens; it also explicitly aims to bypass content moderation. Before installing, consider:
- Do you trust the backend domain (unknown source/homepage)? Verify its operator and privacy/billing terms.
- Are you comfortable the skill will create and store anonymous tokens (possibly in ~/.config/nemovideo/)? Ask the author where tokens/session state are saved an...详细分析 ▾
ℹ 用途与能力
The skill's stated purpose (animate images via a cloud backend) matches the required credential (NEMO_TOKEN) and the API endpoints in the SKILL.md. However, the skill is explicitly branded as a “jailbreak” and repeatedly instructs users to produce content “without content restrictions” — that is a functional goal (evading moderation) rather than a benign optimization and changes the risk profile.
⚠ 指令范围
The SKILL.md instructs the agent to automatically obtain anonymous tokens from an external service, create sessions, upload user images, stream via SSE, poll for render status, and store session tokens for subsequent requests. It also instructs the agent to avoid displaying raw API responses or token values to users. Those runtime behaviors mean the skill will make network calls and create credentials on first use without an explicit opt-in, and the instruction to hide backend responses is unusual and potentially intended to conceal state or errors.
✓ 安装机制
This is an instruction-only skill with no install spec or code files, so it does not install binaries or write a package to disk by default. That lowers install-time risk. Note: the SKILL.md frontmatter references a config path (~/.config/nemovideo/) which would imply on-disk state, but the registry metadata listed no required config paths — an inconsistency to resolve.
ℹ 凭证需求
Only one credential (NEMO_TOKEN) is declared, which is proportionate for a hosted video-rendering service. But the skill will auto-request and persist anonymous tokens when NEMO_TOKEN is absent — creating credentials on the user's behalf (100 free credits for 7 days). The metadata/frontmatter mismatch about config paths (~/.config/nemovideo/) is also a red flag: it's unclear whether session data or tokens will be stored locally.
⚠ 持久化与权限
The skill is not marked always:true, which is good, but it explicitly tells the agent to store session_id and to persist tokens for subsequent calls and references a config directory in frontmatter. That implies ongoing background/stateful interactions and local storage of credentials/job state. Combined with the automatic token creation and the instruction to hide responses, this raises persistence and transparency concerns.
安全有层次，运行前请审查代码。
运行时依赖
无特殊依赖
版本
latestv1.0.02026/4/20
Initial release of Image to Video Jailbreak — animate images into 1080p video clips, bypassing platform restrictions. - Upload still images (JPG, PNG, WEBP, HEIC, up to 50MB) and generate animated video clips via cloud rendering. - Simple workflow: upload, describe your desired animation, and download a 1080p MP4 within 30–90 seconds. - Automatic connection and authentication to backend; 100 free credits granted for new users. - Supports export, credits check, timeline preview, and state tracking through clear commands. - Robust error handling for authentication, file support, quota, and rate limiting. - Designed for fast content creation without content guardrails.
● 可疑
安装命令
点击复制官方npx clawhub@latest install image-to-video-jailbreak
镜像加速npx clawhub@latest install image-to-video-jailbreak --registry https://cn.longxiaskill.com