菜单
Umbrella skill for user document workflows — PDF, Word, Excel, PowerPoint, text extraction, summarization. Dispatches to the right MCP / tool; proposes an install via mcp-discovery when no local capability fits. Adapted from Anthropic's docx/pdf/xlsx/pptx skills.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Fetch and summarise recent papers from arxiv.org. Use this skill when the user wants a digest of today's or this week's cs.AI / cs.LG / cs.CL listings, a focused per-topic roundup, or a quick look at a specific submission.
Lightweight code-review checklist the agent runs before declaring a patch ready. Catches the obvious classes of issue (dead code, missing error handling, weakened tests) without a full security audit. Adapted from addyosmani's "Review" lifecycle phase.
Root-cause-first debugging loop for code or runtime errors: understand the failure before proposing a fix. Adapted from Hermes' systematic- debugging — same iron law, smaller surface, ZLAgent-tool-aware.
Do not declare a task done without running verification. For code, that means running tests / smoke; for answers, that means checking the claim against a fresh source. Adapted from obra/superpowers and addyosmani's "Verify" lifecycle phase.
When the user wants something the agent doesn't have a tool for yet (travel planning, GitHub issues, Linear, weather, filesystem, ...), search the live MCP ecosystem for an existing server, propose installing it, and on confirm wire it up via mcp_manage. Adapted from Hermes' mcporter skill — same flow, ZLAgent-tool-aware.
定期检查 ZLAgent 的 skill、MCP 工具、记忆摘要的变化， 生成一份变更摘要并通过 IM 发送给用户。
基于 SOC 职业分类
| name | document-processing |
| description | Umbrella skill for user document workflows — PDF, Word, Excel, PowerPoint, text extraction, summarization. Dispatches to the right MCP / tool; proposes an install via mcp-discovery when no local capability fits. Adapted from Anthropic's docx/pdf/xlsx/pptx skills. |
| version | 0.1.0 |
| tags | ["knowledge","document","workflow"] |
| metadata | {"hermes":{"created_by":"user"},"zlagent":{"category":"knowledge","triggers":["帮我读 PDF","解析一下这个文件","extract from pdf","read docx","从 word 里取","excel 里","ppt 里","转成 pdf"],"capabilities":["read_url","read_file","mcp_discovery"],"related_skills":["mcp-discovery","web-summary","documentation-and-decisions"]}} |
User sends a document (PDF / Word / Excel / PowerPoint / plain text)
or asks to extract, summarise, or edit one. Also triggers when the
user references a file path or URL ending in .pdf, .docx,
.xlsx, .pptx, .txt, .md.
If the user mentions several files or links at once, treat this as a multi-document request and either process them in sequence or ask which one should go first.
Route each document kind to the right specialised tool. Never invent content that is not in the document.
.txt, .md, .log):
→ use read_file (if local) or read_url (if remote).web-summary skill.mcp__pdf__* or a
plugin tool exposing pdf_read), use it. Otherwise propose
installing an MCP via mcp-discovery (candidate examples:
mcp-server-pdf, pdfplumber-mcp). Never fabricate content by
guessing..docx):
→ similar: mcp__docx__* or propose discovery..xlsx):
→ mcp__xlsx__* or code_execution with openpyxl for small local
files, else propose MCP..pptx):
→ mcp__pptx__* or propose MCP.source: file path (inside workspace) or URL.format: explicit override when the extension lies (common for
URLs without suffix).operation: "extract" | "summarise" | "search" |
"convert".fields (optional): specific fields or pages to target.tool_search with the file type as query.mcp-discovery.文档：<source> (<kind>, <size/pages>)
操作：<operation>
<内容摘要或提取，带页码 / 表格 / 列引用>
来源：p.X-Y / sheet:<name> / slide:<N>
If proposing an install:
这个文档需要 <kind> 支持，本地没有合适的工具。
建议装：<mcp name> — <一句话理由>
装吗？（yes/no）