← 返回 Skills 市场
gawainchin

OpenRouter Vision Agent

作者 gawainchin · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ⚠ suspicious
75
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install openrouter-vision-skill
功能描述
Analyze images using OpenRouter's vision API with x-ai/grok-4.1-fast. Requires OPENROUTER_API_KEY env var or user-provided key. Use when the user asks to des...
安全使用建议
This skill appears to do what it claims (call OpenRouter's vision model) but the registry metadata omits the required OPENROUTER_API_KEY — a clear mismatch. Before installing or providing an API key: (1) Confirm the skill's source and author; prefer skills with a verifiable homepage or source repo. (2) Do not send sensitive/private images — the skill sends images and URLs to openrouter.ai. (3) If you must provide an API key, create a scoped or limited key and be ready to revoke it; never reuse high-privilege or long-lived secrets. (4) Ask the publisher to update the manifest to declare OPENROUTER_API_KEY in requires.env (so platform tooling can enforce least privilege). (5) Test with non-sensitive images first and monitor usage/quotas on your OpenRouter account. If you cannot verify the skill's origin or the manifest is not fixed, avoid supplying secrets or installing it.
功能分析
Type: OpenClaw Skill Name: openrouter-vision-agent Version: 1.0.1 The openrouter-vision-agent skill is a standard implementation for image analysis using the OpenRouter API. It provides clear instructions for using curl to interact with the legitimate endpoint (https://openrouter.ai/api/v1/chat/completions) and includes explicit privacy warnings regarding the transmission of images to the external service. No malicious behaviors, obfuscation, or data exfiltration attempts were found.
能力标签
requires-sensitive-credentials
能力评估
Purpose & Capability
The SKILL.md clearly requires an OPENROUTER_API_KEY and describes sending images to https://openrouter.ai, which is coherent with the name/description. However, the registry metadata lists no required env vars or primary credential. That mismatch (manifest says no env required while runtime instructions require an API key) is an incoherence that should be resolved before trusting the skill.
Instruction Scope
The runtime instructions are narrowly scoped to calling the OpenRouter chat completions endpoint with a model and an image_url payload. They explicitly instruct that images (or URLs) are uploaded to openrouter.ai and warn not to send sensitive images. The instructions do not reference unrelated files, credentials, or system paths.
Install Mechanism
This is an instruction-only skill with no install spec and no code files. That minimizes filesystem risk; nothing is downloaded or installed by the skill itself.
Credentials
The SKILL.md requires a single credential (OPENROUTER_API_KEY) which is appropriate for an API-calling vision skill. However, the registry metadata did not declare this required env var or a primary credential. That discrepancy is concerning because the platform-level manifest should list credentials the skill needs so users and controls can evaluate and protect them.
Persistence & Privilege
The skill is not always-enabled and does not request elevated persistence. It is user-invocable and allowed to run autonomously by default (normal for skills), and there is no indication it modifies other skills or system-wide settings.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install openrouter-vision-skill
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /openrouter-vision-skill 触发
  4. 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
v1.0.1
Declare OPENROUTER_API_KEY as required credential; add privacy warning about sending images to openrouter.ai
v1.0.0
Initial release: image analysis via OpenRouter + Grok 4.1 Fast
元数据
Slug openrouter-vision-skill
版本 1.0.1
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 2
常见问题

OpenRouter Vision Agent 是什么?

Analyze images using OpenRouter's vision API with x-ai/grok-4.1-fast. Requires OPENROUTER_API_KEY env var or user-provided key. Use when the user asks to des... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 75 次。

如何安装 OpenRouter Vision Agent?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install openrouter-vision-skill」即可一键安装,无需额外配置。

OpenRouter Vision Agent 是免费的吗?

是的,OpenRouter Vision Agent 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

OpenRouter Vision Agent 支持哪些平台?

OpenRouter Vision Agent 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 OpenRouter Vision Agent?

由 gawainchin(@gawainchin)开发并维护,当前版本 v1.0.1。

💬 留言讨论