← 返回 Skills 市场
Runwayml
作者
slightly-unrelated
· GitHub ↗
· v1.0.0
· MIT-0
153
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install runwayml
功能描述
Generate AI videos, images, and audio with Runway API. Use when generating video from images, text-to-video, video-to-video, character performance, text-to-i...
安全使用建议
This skill appears to be a straightforward Runway API integration, but the published metadata does not match the SKILL.md. Before installing or enabling it: 1) Verify the skill's source and author identity (the registry shows no homepage). 2) Expect that it will require a Runway API key (RUNWAYML_API_SECRET) and will read local image files if you request image→video—do not provide secrets to untrusted skills. 3) Prefer running in a disposable or sandboxed environment until you trust the skill. 4) Ask the publisher to update the registry metadata to declare the RUNWAYML_API_SECRET env var and any install steps (npm package) so permission and install requirements are transparent. 5) Confirm the API endpoints (dev.runwayml.com) and rate/credit costs in your Runway account before use.
功能分析
Type: OpenClaw Skill
Name: runwayml
Version: 1.0.0
The skill bundle is a standard documentation-based guide for integrating the RunwayML API. It provides legitimate code snippets, cURL examples, and model specifications using official RunwayML domains (api.dev.runwayml.com). No malicious code, data exfiltration, or harmful instructions were identified in SKILL.md or _meta.json.
能力评估
Purpose & Capability
The name and description (Runway video/image/audio generation) match the instructions in SKILL.md: text→video, image→video, and audio examples are provided. However, registry metadata states no required env vars or binaries, while SKILL.md explicitly requires a Runway API key (RUNWAYML_API_SECRET) and the Node.js SDK (@runwayml/sdk). This metadata omission is inconsistent with the stated purpose and usage.
Instruction Scope
SKILL.md gives concrete runtime instructions: install the Node.js SDK, create a client that reads RUNWAYML_API_SECRET, call textToVideo/imageToVideo, and examples that read local files (fs.readFileSync('product.jpg')). These file and env-var accesses are coherent with image→video functionality, but they do instruct the agent to read local files and an API secret—actions that affect local data and credentials and therefore should be declared in metadata.
Install Mechanism
There is no formal install spec in the registry (instruction-only skill). SKILL.md instructs the user/agent to run `npm install @runwayml/sdk` which is a normal dependency for Node usage but is not captured in the package metadata. Lack of an install spec means the agent or integrator may need to install dependencies at runtime; this is an operational gap rather than an immediate code-risk, but it reduces transparency.
Credentials
SKILL.md requires a single API credential (RUNWAYML_API_SECRET) which is proportional to the declared functionality. The concern is that the registry metadata did not declare this required environment variable (metadata lists none). That mismatch could lead to accidental credential exposure or failure to warn users that a secret is needed. The instructions also indicate reading local image files, which is reasonable for image→video but should be explicit in metadata/permissions.
Persistence & Privilege
The skill does not request persistent presence (always: false) and does not include install-time hooks or modifications to other skills. It is user-invocable and allows autonomous invocation (default), which is normal for skills; this alone does not raise extra privilege concerns.
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 runwayml - 安装完成后，直接呼叫该 Skill 的名称或使用
/runwayml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 the RunwayML skill.
- Generate AI video, images, and audio via the Runway API using Node.js SDK or direct REST calls.
- Supports the latest Gen-4.5 model for both text-to-video and image-to-video, plus Gen-4 variants, Google Veo, and ElevenLabs audio models.
- Includes quick start code for major generation modes.
- Provides a model/features table and guidance on ratios, durations, and best prompting practices.
- Async output handling and polling are illustrated.
- Requires a RunwayML API key stored in the environment (`RUNWAYML_API_SECRET`).
元数据
常见问题
Runwayml 是什么？
Generate AI videos, images, and audio with Runway API. Use when generating video from images, text-to-video, video-to-video, character performance, text-to-i... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 153 次。
如何安装 Runwayml？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install runwayml」即可一键安装，无需额外配置。
Runwayml 是免费的吗？
是的，Runwayml 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Runwayml 支持哪些平台？
Runwayml 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 Runwayml？
由 slightly-unrelated（@slightly-unrelated）开发并维护，当前版本 v1.0.0。
推荐 Skills