← 返回 Skills 市场
xuwei-fit2cloud

DataEase

作者 xuwei-fit2cloud · GitHub ↗ · v1.1.0 · MIT-0
cross-platform ⚠ suspicious
142
总下载
1
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install dataease
功能描述
通过自然语言查询 DataEase 组织、切换组织、列出仪表板或数据大屏，并把指定资源导出为截图或 PDF。
使用说明 (SKILL.md)

目标

根据用户的自然语言请求，调用 DataEase 接口完成以下工作：

  • 查询组织列表
  • 切换组织
  • 查询指定组织下的仪表板或大屏列表
  • 导出指定仪表板或大屏的截图或 PDF

输入

用户可能会这样表达需求：

  • 查询当前可用组织
  • 切换到华东组织
  • 查看这个组织下有哪些看板
  • 列出华东组织的大屏
  • 导出“销售总览”看板
  • 截一下“门店运营监控”大屏
  • 导出华东经营分析为 pdf
  • 截销售总览，分辨率 1920*1080，等 5 秒再导出

用户输入中可能包含：

  • 组织 ID
  • 组织名称
  • 资源名称
  • 业务类型：dashboard 或 dataV
  • 输出格式：jpeg 或 pdf
  • 分辨率，例如 1920*1080
  • 额外等待时间

输出

输出内容应包括：

  1. 实际执行的动作
  2. 匹配到的组织或资源
  3. 命中的资源 ID
  4. 实际使用的导出参数
  5. 生成文件本身；如果运行环境不能直接附带文件，则至少返回绝对保存路径
  6. 如果无法准确匹配，则说明原因并返回候选项

执行规则

组织相关

  • 只有用户明确要求“查组织”时，才调用组织树接口
  • 只有用户明确要求“切组织”或已经指明目标组织时，才调用切换组织接口
  • 如果用户只说“查询某组织下的资源”，优先确认是否已给出 orgId
  • 如果只给了组织名称，没有唯一命中的组织，不要猜测，应返回候选组织

资源相关

  • 如果用户明确提到“大屏”“数据大屏”“驾驶舱”，则使用 dataV
  • 其他情况默认使用 dashboard
  • 查询资源时优先列出叶子节点
  • 如果用户只是想“看看有哪些资源”，返回资源列表
  • 如果用户要求“导出”或“查看”某个资源，先匹配资源，再执行本地预览页截图流程
  • 当存在多个相似候选时，不要直接猜测，应返回候选列表并说明歧义

导出默认值

  • busiType=dashboard
  • pixel=1920*1080
  • extWaitTime=0
  • resultFormat=0，表示 JPEG
  • 如果用户明确要求 PDF，则使用 resultFormat=1

名称匹配规则

在匹配组织名称或资源名称前，需要做标准化处理：

  • 去掉前后空格
  • 英文统一转小写
  • 忽略中英文引号差异
  • 忽略常见标点差异
  • 资源优先匹配叶子节点
  • 如果存在别名映射，先用别名映射转换，再做匹配

鉴权与安全

  • 不要把 access key、secret key 等敏感凭证硬编码到 skill 文件中
  • 不要要求用户在提问里携带 base_url、access key、secret key、username、password
  • 本技能通过以下配置读取环境：
    • DATAEASE_BASE_URL
    • DATAEASE_ACCESS_KEY
    • DATAEASE_SECRET_KEY
    • DATAEASE_USERNAME
    • DATAEASE_PASSWORD
    • DATAEASE_LOGIN_ORIGIN
  • 脚本会自动读取仓库根目录 .env
  • 如果提供了 username 和 password，脚本会先调用 dekey 和 /login/localLogin 获取 x-de-token
  • 如果提供了 accessKey 和 secretKey，脚本会自动生成：
    • signature
    • x-de-ask-token
  • 如果已切换组织，则后续查询资源或导出时使用切组织返回的 x-de-token
  • 如果用户已提供现成的 x-de-token，脚本可直接复用

失败处理

  • 如果组织树接口调用失败，明确说明组织列表加载失败
  • 如果切换组织失败，返回 HTTP 状态和响应内容
  • 如果资源树接口调用失败，明确说明资源列表加载失败
  • 如果没有找到合适资源，返回最接近的候选名称
  • 如果本地浏览器启动失败、预览页加载失败或截图失败，明确返回错误原因
  • 只有在文件实际写入成功后，才能声明导出成功
  • 导出成功后必须返回 saved_file

使用文件

  • 使用 scripts/capture_dashboard.py 执行 API 调用和文件保存
  • 使用 scripts/browser_capture.mjs 打开预览页并完成本地截图
  • 使用 references/api.md 查看接口说明和鉴权接入方式
  • 如需维护资源别名，编辑 references/resource_aliases.json

推荐命令

  • 查询组织：
    • python3 scripts/capture_dashboard.py list-orgs
  • 切换组织：
    • python3 scripts/capture_dashboard.py switch-org --org-id 1225813472202330112
  • 查询指定组织下的看板：
    • python3 scripts/capture_dashboard.py list-resources --org-id 1225813472202330112 --busi-type dashboard
  • 查询指定组织下的大屏：
    • python3 scripts/capture_dashboard.py list-resources --org-id 1225813472202330112 --busi-type dataV
  • 导出看板截图：
    • python3 scripts/capture_dashboard.py capture --org-id 1225813472202330112 --resource-name 销售总览 --busi-type dashboard
  • 导出大屏 PDF：
    • python3 scripts/capture_dashboard.py capture --org-id 1225813472202330112 --resource-name 门店运营监控 --busi-type dataV --result-format 1
安全使用建议
This skill appears to implement what it claims, but there are notable mismatches and runtime behaviors you should consider before installing or letting an agent invoke it autonomously: - Credentials and .env: The code expects DATAEASE_BASE_URL and either (DATAEASE_ACCESS_KEY + DATAEASE_SECRET_KEY) or (DATAEASE_USERNAME + DATAEASE_PASSWORD) and may read a .env in the repo root. The registry metadata did not declare these — do not paste credentials into chat. Provide credentials only via environment variables or ephemeral tokens and prefer least-privilege tokens. - Binaries and dependencies: The skill requires Python3, Node.js, Playwright (chromium), and the system openssl binary. There is no install spec, so you must run the README steps (npm install, npx playwright install chromium) manually in a safe environment. If you cannot or do not want to install these, do not enable the skill. - Data captured: The browser capture collects slices of page text and HTML for diagnostics (bodyText, appHtml) and injects tokens into localStorage to authenticate the preview page. These behaviors are necessary for rendering/export but may expose sensitive content from the preview page — run the skill only in a controlled environment. - Safest approach: Review the scripts yourself (they are included), run them manually in an isolated environment with ephemeral or scoped credentials, and confirm expected behavior before allowing the agent to invoke the skill autonomously. Request that the skill publisher update the manifest to declare required env vars and required binaries/install steps before trusting automated installation.
功能分析
Type: OpenClaw Skill Name: dataease Version: 1.1.0 The skill bundle provides a legitimate integration for DataEase, enabling users to query organizations and export dashboards or data screens as images or PDFs. The implementation uses a Python script (capture_dashboard.py) for API orchestration and a Node.js script (browser_capture.mjs) using Playwright for rendering and capturing the visual resources. Authentication is handled securely via environment variables or .env files, supporting both API key signatures and username/password flows with appropriate RSA/AES encryption. While the scripts utilize system calls to 'openssl' and 'node', they employ safe subprocess execution patterns with sanitized arguments. No evidence of data exfiltration, persistence, or malicious prompt injection was found.
能力评估
⚠ Purpose & Capability
The skill's name/description (export DataEase dashboards) aligns with the included Python and Node scripts which call DataEase APIs and drive a headless browser. However the registry metadata declares no required environment variables or binaries while SKILL.md and the code clearly expect multiple environment variables (DATAEASE_BASE_URL, DATAEASE_ACCESS_KEY, DATAEASE_SECRET_KEY, DATAEASE_USERNAME, DATAEASE_PASSWORD, DATAEASE_LOGIN_ORIGIN) and external tools (openssl, Python3, Node/Playwright, Chromium). The omission in the manifest is an incoherence that could mislead users or automated installers.
ℹ Instruction Scope
SKILL.md instructions are scoped to listing orgs, switching orgs, listing resources, and exporting via a local browser preview; they instruct reading environment variables and .env, building tokens, calling DataEase endpoints, and running local browser capture scripts. The capture script's diagnostics collect page body/app HTML snippets and visible text (used for debugging) which could surface sensitive data from the preview page; while useful for diagnosing render issues, that behavior should be noted as it reads page content beyond just capturing an image/PDF.
⚠ Install Mechanism
There is no formal install specification in the registry despite a package.json and README that require npm install, Playwright installation, and Python dependencies. The code relies on native openssl via subprocess and on Playwright/Chromium for headless capture; absence of an install step means an agent or user might run the skill in an environment lacking these components, causing failures or prompting ad-hoc installation actions. Missing install metadata is both a usability and a security concern (unexpected binary usage).
⚠ Credentials
The skill needs sensitive configuration (access keys or username/password and base URL) which are proportionate to interacting with DataEase, but the skill manifest did not declare any required environment variables. The code will automatically read a .env in the repository root and will accept either credential method (accessKey+secretKey or username+password) and can consume an existing x-de-token. That the manifest omits these expectations is a coherence problem; users should treat provided secrets as high-value and avoid embedding them in chat messages.
✓ Persistence & Privilege
The skill does not request 'always: true' and does not claim to modify other skills or global agent settings. It writes outputs (saved files) and may create temporary files, which is expected for an export tool. Autonomous invocation is allowed (platform default), so be cautious about giving it live credentials if you plan to let agents call it without supervision.
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 dataease
  3. 安装完成后，直接呼叫该 Skill 的名称或使用 /dataease 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
v1.1.0
Version 1.1.0 - 增加对用户名密码（`username`、`password`）鉴权方式的支持，可自动调用 `dekey` 和 `/login/localLogin` 获取 `x-de-token` - 新增支持通过现成的 `x-de-token` 进行复用，无需重复鉴权 - 新增环境变量配置项：`DATAEASE_USERNAME`、`DATAEASE_PASSWORD`、`DATAEASE_LOGIN_ORIGIN` - 明确禁止用户在提问中携带敏感凭证信息 - 优化了凭证优先级和切组织后 token 使用方式说明
v1.0.0
dataease-resource-skill v1.0.0 — Initial release. - 支持用自然语言查询 DataEase 组织、切换组织、列出仪表板/大屏。 - 可导出指定仪表板或大屏为截图（JPEG）或 PDF 文件。 - 智能识别组织和资源名称，并提供歧义处理和候选建议。 - 自动读取环境配置，保障鉴权安全。 - 明确失败场景，详细错误说明，成功导出返回绝对文件路径。
元数据
Slug dataease
版本 1.1.0
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

DataEase 是什么？

通过自然语言查询 DataEase 组织、切换组织、列出仪表板或数据大屏，并把指定资源导出为截图或 PDF。 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 DataEase？

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

DataEase 是免费的吗？

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

DataEase 支持哪些平台？

DataEase 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 DataEase？

由 xuwei-fit2cloud（@xuwei-fit2cloud）开发并维护，当前版本 v1.1.0。

推荐 Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 留言讨论