← 返回 Skills 市场
142
总下载
0
收藏
0
当前安装
6
版本数
在 OpenClaw 中安装
/install paperbanana-dashscope
功能描述
Generate academic figures and scientific diagrams from paper text using a multi-agent pipeline powered by Alibaba Cloud DashScope (Qwen-VL + Wanxiang/Qwen-Im...
安全使用建议
Proceed cautiously. Before installing or providing any API key: 1) Inspect the npm package and the linked GitHub repository (look at package.json, install/postinstall scripts, and the code that sends requests). 2) Do not reuse your primary OpenAI API key — create a dedicated key with minimal privileges or use a throwaway/test key. 3) Verify whether DashScope truly accepts an OpenAI-style key or if the package is repurposing your OpenAI key to talk to other services. 4) Prefer installing in a sandbox/container or examining the package source locally instead of installing globally. 5) If you cannot confirm why OPENAI_API_KEY is required for an Alibaba service, contact the package author or avoid installing.
功能分析
Type: OpenClaw Skill
Name: paperbanana-dashscope
Version: 1.0.5
The skill bundle describes a legitimate CLI tool for generating academic figures using Alibaba Cloud's DashScope API. The instructions in SKILL.md and metadata in clawhub.json align with the stated purpose of research visualization, using standard configuration patterns (e.g., ~/.paperbanana-dashscope/config.yaml) and providing links to public npm and GitHub repositories.
能力评估
Purpose & Capability
The skill claims to use Alibaba Cloud DashScope (Qwen-VL / Wanxiang / Qwen-Image) to generate academic figures, which matches the stated purpose. However, both SKILL.md and clawhub.json require an OPENAI_API_KEY (and the sample config uses openai_api_key). Requiring an OpenAI-style key for an Alibaba DashScope integration is unexpected and not explained in the docs — this mismatch is incoherent and worth verifying with the package author or upstream repo.
Instruction Scope
SKILL.md is an instruction-only CLI guide: it tells the user to install an npm package globally and to set an API key (OPENAI_API_KEY) or create a ~/.paperbanana-dashscope config. The instructions do not explicitly ask the agent to read unrelated system files or exfiltrate data. However, the runtime behavior depends entirely on the npm package (not included here), so the instructions alone cannot guarantee what the installed tool will do.
Install Mechanism
There is no install spec in the skill bundle itself (instruction-only). The README recommends npm install -g paperbanana-dashscope from npmjs.com and links a GitHub repo. Using npm and GitHub is a common install path, but global npm installation runs code on the host and may execute install scripts; you should inspect the package on npm/github before installing.
Credentials
Only OPENAI_API_KEY (required) and an optional DASHSCOPE_IMAGE_MODEL are declared. The required credential name (OPENAI_API_KEY) does not clearly match the declared provider (Alibaba DashScope). This raises two concerns: (1) the package may be asking users to reuse sensitive OpenAI keys for an unrelated third-party service, and (2) it may send that key to endpoints outside OpenAI. No other unrelated credentials are requested, but the mismatch is disproportionate and unexplained.
Persistence & Privilege
The skill does not request always: true, does not include install scripts in the skill bundle, and declares no required system config paths. It's user-invocable and allows model invocation (normal). There is no evidence the skill modifies global agent settings itself.
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 paperbanana-dashscope - 安装完成后，直接呼叫该 Skill 的名称或使用
/paperbanana-dashscope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
v1.0.5
新增 wan2.7-image 和 wan2.7-image-pro 支持。wan2.7-image 与 wan2.6 同价但提示词容量提升至 5000 字符，wan2.7-image-pro 支持 4K 输出。
v1.0.4
完全重写为原生 TypeScript CLI，零 Python 依赖。一条命令安装：npm install -g paperbanana-dashscope。新增 wan2.6-t2i / qwen-image-plus 等新模型支持。
v1.0.3
v1.0.3 (No code or documentation changes detected)
- No changes since the previous version.
- All usage, parameters, and setup remain the same.
v1.0.2
Version 1.0.2 of paperbanana-dashscope
- No code or documentation changes detected in this version.
- The SKILL.md and all implementation files remain unchanged from the previous release.
v1.0.1
首次发布。基于 PaperBanana/PaperVizAgent 的阿里云 DashScope 适配版，使用 qwen-vl-max 作为 VLM，wanx2.1-t2i-turbo 作为图像生成模型。支持从论文方法章节自动生成学术插图。
v1.0.0
首次发布。基于 PaperBanana/PaperVizAgent 的阿里云 DashScope 适配版，使用 qwen-vl-max 作为 VLM，wanx2.1-t2i-turbo 作为图像生成模型。支持从论文方法章节自动生成学术插图。
元数据
常见问题
Clawhub Skill 是什么？
Generate academic figures and scientific diagrams from paper text using a multi-agent pipeline powered by Alibaba Cloud DashScope (Qwen-VL + Wanxiang/Qwen-Im... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 142 次。
如何安装 Clawhub Skill？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install paperbanana-dashscope」即可一键安装，无需额外配置。
Clawhub Skill 是免费的吗？
是的，Clawhub Skill 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Clawhub Skill 支持哪些平台？
Clawhub Skill 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 Clawhub Skill？
由 yrjm（@zdywrnm）开发并维护，当前版本 v1.0.5。
推荐 Skills