← 返回 Skills 市场
3937
总下载
8
收藏
31
当前安装
1
版本数
在 OpenClaw 中安装
/install flyworks-avatar-video
功能描述
Generate videos using Flyworks (a.k.a HiFly) Digital Humans. Create talking photo videos from images, use public avatars with TTS, or clone voices for custom audio.
安全使用建议
Review before installing. Use your own Flyworks/HiFly token, treat it as a secret, and assume any local image or audio path you provide may be uploaded for remote processing. Only upload portraits and voice samples you own or have explicit permission to use, and check the provider's retention, deletion, and acceptable-use terms before processing sensitive media.
功能分析
Type: OpenClaw Skill
Name: flyworks-avatar-video
Version: 1.0.0
The skill is designed to interact with the Flyworks API for video generation, avatar creation, and voice cloning. It uses `requests` to communicate with `hfw-api.hifly.cc` and reads user-specified local image/audio files for upload, which is a core function of the skill. The `SKILL.md` contains agent instructions that are benign and directly aligned with the stated purpose. There is no evidence of data exfiltration beyond user-provided files, malicious execution, persistence mechanisms, or prompt injection attempts with harmful objectives.
能力评估
Purpose & Capability
The observed code and documentation match the stated purpose: it calls the Flyworks/HiFly API to list avatars and voices, create videos, create talking-photo avatars, clone voices, and upload user-provided image or audio inputs when needed.
Instruction Scope
The skill gives operational instructions and examples for voice cloning and talking-photo generation, but does not clearly require consent, warn about impersonation risk, or explain the sensitivity of face and voice data.
Install Mechanism
Installation uses normal clone/copy/add-skill and pip dependency steps, with only the requests package listed; no hidden installer, obfuscation, or destructive install behavior was found.
Credentials
Network access, environment token use, and local media file reads are purpose-aligned, but local image and audio files are uploaded to a third-party service without an explicit confirmation step or clear data-handling disclosure at the point of use.
Persistence & Privilege
The script writes a local memory.json alias file for saved avatar and voice IDs and embeds a fallback demo API token; there is no background persistence or privilege escalation, but the shared hardcoded token and saved generated-identity references deserve review.
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 flyworks-avatar-video - 安装完成后，直接呼叫该 Skill 的名称或使用
/flyworks-avatar-video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 flyworks-avatar-video skill.
- Generate videos using Flyworks/HiFly Digital Humans: talking photos, public avatars with TTS, or cloned voices.
- Includes command-line tool for listing avatars/voices, creating videos, talking photo videos, and voice cloning.
- Supports workflow with free-tier token (with watermark and 30s video limit); instructions provided for setting up custom API keys.
- Provides agent guidelines for interactive user support and memory management of avatars/voices.
元数据
常见问题
Flyworks Avatar Video 是什么？
Generate videos using Flyworks (a.k.a HiFly) Digital Humans. Create talking photo videos from images, use public avatars with TTS, or clone voices for custom audio. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 3937 次。
如何安装 Flyworks Avatar Video？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install flyworks-avatar-video」即可一键安装，无需额外配置。
Flyworks Avatar Video 是免费的吗？
是的，Flyworks Avatar Video 完全免费（开源免费），可自由下载、安装和使用。
Flyworks Avatar Video 支持哪些平台？
Flyworks Avatar Video 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 Flyworks Avatar Video？
由 Hui Lin（@linhui99）开发并维护，当前版本 v1.0.0。
推荐 Skills