docx-chapterResearch-backed chapter pipeline: outline → parallel ACP research → MD synthesis → DOCX generation → source verification → screenshot verification (mandatory)
Install via ClawdBot CLI:
clawdbot install ccmxigua/docx-chapterGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Calls external URL not in known-safe list
https://github.com/anthropics/claude-code/issues/50597Audited Jun 5, 2026 · audit v1.0
Generated Oct 6, 2026
A consulting firm needs to produce market research whitepapers with dozens of chapters. This skill turns an outlined chapter into a fully researched, cited, and formatted DOCX with verified sources and screenshot-backed claims, saving analysts days of manual compilation. Final delivery includes footnote integrity checks and deduplication reports.
Government affairs teams must generate chapters covering regulations backed by authoritative sources (gov/academic). The skill's dual-search pipeline (unified_search + Tavily) and source verification stage ensure every claim traces to a real, accessible URL. Reference DOCX support allows enforcing corporate or official style templates.
Researchers compile chapter-length literature reviews with rigorous citations. The chapter.json and research_results.json pipeline plus claim-driven keyword generation produces English footnotes for Chinese-language drafts, enabling screenshot-level verification that cited content actually appears on the source page.
Marketing teams need up-to-date competitive intel packaged as chaptered reports. Parallel ACP research enables coverage of many sub-topics at once, while the URL dedup script keeps footnotes clean. The 30% empty-result failure mode is mitigated by automatic retries and unified_search fallback.
Investment research desks require sourced data points (metrics, quotes) with verifiable citations. The verification_keywords.json artifact supports screenshot verification for each numeric claim, while the inline-citation scanner prevents stray (Author, 2023) references from leaking into the DOCX.
Charge research firms or enterprises per generated chapter/document, priced by section count and citation depth. Suitable for occasional use without long-term contracts.
Monthly subscription that bundles the skill with unified-search and Tavily-research dependencies, hosted infrastructure for ACP execution, and version updates. Targets analysts who produce chapters weekly.
Charge for a managed service where the vendor runs the pipeline end-to-end (outline review, parallel research, DOCX generation, verification) and delivers branded chapters to the client. Higher touch, includes quality assurance on failed ACP runs.
💬 Integration Tip
Install required skills (unified-search, tavily-research) and set the Tavily API key plus pandoc before first run; be aware that ACP Claude CLI has a ~30% empty-result rate requiring retries and unified_search fallback, and that the cross-agent URL dedup script MUST run before chapter.md generation to keep footnotes valid.
Scored Oct 6, 2026
PDF智能处理工具 v2.1 | PDF Smart Tool. 支持PDF转换、OCR识别、合并拆分、数字签名、批量处理、水印添加、加密解密。触发词：PDF、转换、识别。
Generate hand-drawn style diagrams, flowcharts, and architecture diagrams as PNG images from Excalidraw JSON
Convert public web pages into clean Markdown with markdown.new for AI workflows. Use when tasks require URL-to-Markdown conversion for summarization, RAG ing...
PDF扫描件转Word文档。支持中文OCR识别，自动裁掉页眉页脚，保留插图，彩色章节封面页保留为图片。使用百度OCR API（免费额度1000次/月）。当用户要求把扫描PDF转成文字/Word时触发。
Word文档处理工具套件，提供Word文档的创建、读取、内容提取和基本处理功能。
Microsoft SharePoint and OneDrive integration with managed OAuth. Manage sites, lists, libraries, files, folders, permissions, content types, and SharePoint...