← 返回 Skills 市场
Pika Ai Video Generator
作者
francemichaell-15
· GitHub ↗
· v1.0.0
· MIT-0
72
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install pika-ai-video-generator
功能描述
Cloud-based pika-ai-video-generator tool that handles generating short AI videos from text or image prompts. Upload JPG, PNG, MP4, MOV files (up to 200MB), d...
安全使用建议
This skill appears to be a legitimate cloud video generator, but there are some things to check before installing/using it: 1) The skill will auto-create an anonymous token if you don't provide NEMO_TOKEN — ask where that token and session_id are stored and how long they're valid. 2) The runtime tells the agent to hide raw API responses and token values; lack of visible responses can be used to obscure unexpected behavior. 3) The skill adds headers that reveal install-path-derived platform info and may request file uploads by local path — ensure you only upload intended media and not sensitive filesystem paths. 4) The source is unknown and homepage is missing; prefer skills from verified authors or ask the publisher for provenance and a privacy/data-retention policy. If you proceed, provide your own NEMO_TOKEN (ephemeral/scoped if possible), avoid uploading sensitive files, and ask the maintainer to clarify storage location for session tokens and the reason for the configPath/frontmatter discrepancy.
功能分析
Type: OpenClaw Skill
Name: pika-ai-video-generator
Version: 1.0.0
The skill bundle exhibits impersonation and data-routing concerns by using 'Pika AI' branding while directing all traffic, including user-uploaded files and prompts, to an unrelated third-party domain (mega-api-prod.nemovideo.ai). It instructs the AI agent to perform automated background connections and environment fingerprinting (leaking the agent's install path via the 'X-Skill-Platform' header) while explicitly directing the agent to hide raw API responses and tokens from the user. These behaviors, found in SKILL.md, are characteristic of unauthorized data collection or telemetry wrappers.
能力评估
Purpose & Capability
The declared purpose (cloud AI video generation) aligns with the API endpoints and the single required credential (NEMO_TOKEN). However, the SKILL.md frontmatter declares a config path (~/.config/nemovideo/) while the registry metadata lists no required config paths — a minor inconsistency in what the skill claims it needs.
Instruction Scope
Runtime instructions include auto-creating anonymous tokens if NEMO_TOKEN is not present, storing a session_id for later use, and explicitly instruct the agent not to display raw API responses or token values. The skill also instructs uploading files by local path (multipart -F "files=@/path"), and adds headers that leak install-path-derived platform info. These behaviors broaden the agent's actions (creating credentials, storing session state, reading local file paths, and sending environment/hardware-fingerprint-like headers) beyond a minimal request/response workflow and are not fully justified or explained.
Install Mechanism
Instruction-only skill with no install spec or bundled code; nothing is written to disk by an installer. This is the lowest install risk.
Credentials
Only one credential (NEMO_TOKEN) is declared, which is proportionate for a cloud API. But the SKILL.md instructs generating and storing anonymous tokens when the env var is absent and includes a configPaths entry in frontmatter — the storage location and lifetime of session/token are unspecified, and the automatic creation/storage behavior should be clarified before trust.
Persistence & Privilege
always:false and normal autonomous invocation. The skill instructs an automatic first-time backend connection and to 'store the returned session_id' but does not say where; this raises a small persistence/privilege question (where/how long session data is kept). No explicit requests to modify other skills or system-wide config were found.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install pika-ai-video-generator - 安装完成后，直接呼叫该 Skill 的名称或使用
/pika-ai-video-generator触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release of Pika AI Video Generator — Generate AI Videos from Prompts
- Generate 1080p MP4 videos from text or image prompts in 30–90 seconds.
- Supports uploads of JPG, PNG, MP4, and MOV files up to 200MB.
- Handles automatic connection, authentication, and session management with the backend.
- Provides workflows for tracking credits, session state, exporting videos, and previewing timelines.
- Includes robust error handling and clear guidance for common issues and best results.
元数据
常见问题
Pika Ai Video Generator 是什么？
Cloud-based pika-ai-video-generator tool that handles generating short AI videos from text or image prompts. Upload JPG, PNG, MP4, MOV files (up to 200MB), d... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 72 次。
如何安装 Pika Ai Video Generator？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install pika-ai-video-generator」即可一键安装，无需额外配置。
Pika Ai Video Generator 是免费的吗？
是的，Pika Ai Video Generator 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Pika Ai Video Generator 支持哪些平台？
Pika Ai Video Generator 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Pika Ai Video Generator？
由 francemichaell-15（@francemichaell-15）开发并维护，当前版本 v1.0.0。
推荐 Skills