← 返回 Skills 市场
125
总下载
1
收藏
0
当前安装
4
版本数
在 OpenClaw 中安装
/install pdflux-test
功能描述
Convert unstructured documents into LLM-ready structured data. Supports PDF, Word, PPT, and images; extracts paragraphs, formulas, tables, charts, and other...
安全使用建议
This skill will upload any local file you give it to an external SaaS endpoint (https://platform.paodingai.com) using the PD_ROUTER_API_KEY you supply—do not provide that key unless you trust the service. Avoid uploading sensitive or confidential documents unless you have permission and have reviewed the service's privacy/retention policies. Confirm you run the bundled script path present in the package (scripts/upload_to_markdown.js) and ensure your Node runtime supports global fetch/FormData if running locally. If you want to prevent accidental automatic uploads, do not enable autonomous agent runs that might provide files without explicit human confirmation.
功能分析
Type: OpenClaw Skill
Name: pdflux-test
Version: 1.0.3
The skill is a legitimate wrapper for the PDFlux document conversion service. It provides a Node.js script (upload_to_markdown.js) that uploads a user-specified file to the official PaodingAI platform, polls for processing status, and retrieves the resulting Markdown. The code uses standard Node.js APIs, follows the stated purpose in SKILL.md, and does not exhibit signs of data exfiltration to unauthorized endpoints or malicious execution.
能力评估
Purpose & Capability
Name/description match the actual behavior: the script uploads a local file to https://platform.paodingai.com via the pdflux service, polls parsing status, and downloads Markdown. Required binary (node) and primary env var (PD_ROUTER_API_KEY) are appropriate.
Instruction Scope
SKILL.md instructs the agent to run the bundled script and to ask for PD_ROUTER_API_KEY if missing. The instructions do not ask the agent to read unrelated files, keys, or system state, nor to send data to unexpected endpoints.
Install Mechanism
No external install/remote download is performed by the skill; the functionality is entirely in the included script (instruction-only install). No suspicious third-party URLs or archive extraction are used.
Credentials
Only one required credential (PD_ROUTER_API_KEY) is declared and used solely to authenticate requests to the stated endpoint. Optional env vars (PDFLUX_*) control parsing behavior and are relevant. No unrelated secrets or config paths are requested.
Persistence & Privilege
Skill is not always-included and does not request system-wide persistence or modify other skills. Autonomous invocation is allowed (platform default) but is not combined with other privilege escalation indicators.
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 pdflux-test - 安装完成后，直接呼叫该 Skill 的名称或使用
/pdflux-test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
v1.0.3
- Added explicit security and data flow section to SKILL.md, clarifying upload endpoint, authentication, and no support for arbitrary hosts.
- Updated required environment variables to only mandate PD_ROUTER_API_KEY; others are now optional.
- Changed metadata to specify "primaryEnv" as PD_ROUTER_API_KEY.
- Removed mention of configurable upload host in environment variables.
- Clarified documentation about official endpoints and limited network scope.
- Bumped version metadata to 1.0.2 in SKILL.md.
v1.0.2
- Added support for the PD_ROUTER_BASE_URL environment variable, allowing optional specification of the PDRouter service address.
- Updated documentation to include PD_ROUTER_BASE_URL as an optional environment variable.
v1.0.1
- Updated documentation to use English throughout for broader accessibility.
- Revised metadata: changed author to "PAODINGAI", updated version to 1.0.1, and added new required environment variables (`PD_ROUTER_SERVICE_CODE`, `PD_ROUTER_BASE_URL`).
- Standardized environment variables (`PD_ROUTER_API_KEY`, `PDFLUX_INCLUDE_IMAGES`) in documentation and metadata.
- Improved and clarified usage instructions, execution constraints, and scenarios for when to use the skill.
- Enhanced section descriptions for default behaviors, optional parameters, and script workflow.
v1.0.0
PDFlux-PDF2Markdown 1.0.0 初始发布
- 支持一键将 PDF、Word、PPT、图片等文档解析为多级目录结构的 Markdown。
- 自动提取段落、公式、表格、图表等元素，便于后续知识检索、字段提取或智能问答。
- 上传文件后自动轮询解析状态，并在完成后下载输出 Markdown。
- 提供灵活的图片/图表包含选项，默认不含图片，可通过环境变量启用。
- 明确约定调用解析脚本的流程及适用场景，以确保一致性和高质量输出。
- 环境变量控制鉴权与图片输出，缺失时会有清晰的报错与提示。
元数据
常见问题
PDFlux Test 是什么？
Convert unstructured documents into LLM-ready structured data. Supports PDF, Word, PPT, and images; extracts paragraphs, formulas, tables, charts, and other... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 125 次。
如何安装 PDFlux Test？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install pdflux-test」即可一键安装，无需额外配置。
PDFlux Test 是免费的吗？
是的，PDFlux Test 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
PDFlux Test 支持哪些平台？
PDFlux Test 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 PDFlux Test？
由 Sube（@sube-py）开发并维护，当前版本 v1.0.3。
推荐 Skills