← 返回 Skills 市场
4939
总下载
1
收藏
24
当前安装
1
版本数
在 OpenClaw 中安装
/install fal-ai
功能描述
Generate images, videos, and audio via fal.ai API (FLUX, SDXL, Whisper, etc.)
安全使用建议
Install only if you are comfortable giving the skill a fal.ai API key and sending prompts, image URLs, audio URLs, and generation parameters to fal.ai. Use a dedicated or revocable key where possible, monitor credit usage, and avoid submitting secrets, personal data, or regulated content unless fal.ai processing is acceptable for your use case.
功能分析
Type: OpenClaw Skill
Name: fal-ai
Version: 0.1.0
The skill is designed to interact with the fal.ai API for generative media. It retrieves the API key from environment variables or `clawdbot` configuration, which is a standard and expected practice for OpenClaw skills. All network requests are directed to the legitimate `https://queue.fal.run` endpoint, and user inputs like prompts are sent as data to this API, not executed locally. The use of `subprocess.run` in `fal_api.py` is limited to safely retrieving configuration from `clawdbot` and does not pose a command injection risk. There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection against the agent.
能力评估
Purpose & Capability
The capability matches the stated purpose: the Python helper submits image, video, and audio jobs to fal.ai's queue API and can target several documented model endpoints; this may consume fal.ai quota or credits.
Instruction Scope
The instructions are setup and usage documentation only, with no hidden agent-control, prompt override, unsafe automatic execution, or unrelated behavior.
Install Mechanism
No dependency installation or installer script is present; the package contains markdown metadata and a stdlib-only Python helper.
Credentials
The requested FAL_KEY and outbound calls to https://queue.fal.run are proportionate to the API-client purpose, but prompts, media URLs, and related payload data leave the local environment.
Persistence & Privilege
No persistence, background worker, privilege escalation, destructive file access, or broad local indexing was found; subprocess use is limited to a fixed clawdbot config lookup.
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 fal-ai - 安装完成后，直接呼叫该 Skill 的名称或使用
/fal-ai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
v0.1.0
fal-ai 0.1.0 changelog:
- Initial release of the fal.ai API skill.
- Supports generation of images (FLUX, SDXL, Recraft), videos (MiniMax, WAN), and audio transcripts (Whisper).
- Async queue-based generation with simple submit–poll–result workflow.
- 600+ model endpoints supported; easy listing of available models.
- No external dependencies required (stdlib only, no fal_client).
- API key setup via environment variable or config command.
元数据
常见问题
Fal.ai API 是什么？
Generate images, videos, and audio via fal.ai API (FLUX, SDXL, Whisper, etc.). 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 4939 次。
如何安装 Fal.ai API？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install fal-ai」即可一键安装，无需额外配置。
Fal.ai API 是免费的吗？
是的，Fal.ai API 完全免费（开源免费），可自由下载、安装和使用。
Fal.ai API 支持哪些平台？
Fal.ai API 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 Fal.ai API？
由 agmmnn（@agmmnn）开发并维护，当前版本 v0.1.0。
推荐 Skills