← 返回 Skills 市场
557
总下载
2
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install vajra
功能描述
Analyze URLs, YouTube videos, tweets, or text for quality, bias, and reliability using the Vajra API (vajra.to). Use when the user asks to fact-check, analyz...
安全使用建议
This skill appears coherent with its stated purpose, but note these important privacy considerations before installing: any content you submit is transmitted to vajra.to for analysis; analyses are cached and a public permalink is created (so do not submit private, proprietary, or PII-containing material); the public retrieve endpoint means anyone with the analysis ID can view the report; and the free tier limits analyses per month. If you must analyze sensitive content, do not use this service. Otherwise, providing the single VAJRA_API_KEY is expected and proportional.
功能分析
Type: OpenClaw Skill
Name: vajra
Version: 1.1.0
The skill is designed to analyze content via the Vajra API, explicitly stating that user-provided content is sent to `vajra.to`, cached, and results in a public permalink. It transparently warns users not to submit private data. While the `curl` examples could pose a shell injection vulnerability if the AI agent directly executes them with unsanitized user input, this represents a potential flaw in the agent's execution model rather than malicious intent within the skill itself. There is no evidence of data exfiltration beyond the stated purpose, malicious execution, persistence, or deceptive prompt injection against the agent.
能力评估
Purpose & Capability
Name/description (content analysis, bias/reliability scoring) match the runtime instructions and the single required env var (VAJRA_API_KEY). Requiring an API key for an external analysis API is expected.
Instruction Scope
SKILL.md only instructs sending content (URLs/text/video/tweets) to vajra.to and formatting the returned JSON. It explicitly warns about caching and public permalinks and does not instruct the agent to read unrelated files, other env vars, or system configuration.
Install Mechanism
No install spec and no code files — instruction-only skill. This is the lowest-risk install model and consistent with the described curl usage.
Credentials
Only VAJRA_API_KEY is required. That single credential is appropriate for an external API. No unrelated secrets, config paths, or multiple credentials are requested.
Persistence & Privilege
The skill does not request always:true and does not ask to modify other skills or system settings. It requires the API key only for its own use.
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 vajra - 安装完成后，直接呼叫该 Skill 的名称或使用
/vajra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
v1.1.0
- Added detailed privacy and data handling information, including caching and public permalink behavior.
- Clarified setup instructions with explicit environment variable requirements.
- Revised and expanded documentation for supported content types and limitations.
- Improved usage instructions for presenting results and fetching prior analyses.
- Added metadata section specifying required environment variables.
v1.0.0
- Initial release of the Vajra skill.
- Analyze URLs, YouTube videos, tweets, or raw text for quality, bias, and reliability using the Vajra API.
- Returns a structured verdict, quality score (1–10), bias assessment, key takeaways, and warnings for the provided content.
- Provides shareable permalinks to full analysis reports.
- Requires API key—supports authentication via environment variable or user input.
- Supports cached results at no credit cost and includes usage information for free and paid plans.
元数据
常见问题
Vajra 是什么？
Analyze URLs, YouTube videos, tweets, or text for quality, bias, and reliability using the Vajra API (vajra.to). Use when the user asks to fact-check, analyz... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 557 次。
如何安装 Vajra？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install vajra」即可一键安装，无需额外配置。
Vajra 是免费的吗？
是的，Vajra 完全免费（开源免费），可自由下载、安装和使用。
Vajra 支持哪些平台？
Vajra 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 Vajra？
由 Mig（@minhyeong112）开发并维护，当前版本 v1.1.0。
推荐 Skills