商业航天文档对齐工作台：面向航天工程团队的跨部门追溯、草稿生成与变更影响分析。
Query and validate Markdown as a document graph—front matter, links, policies, change impact, CLI, and MCP for AI agents.
Structure-aware PDF chunking for RAG. It builds a document hierarchy from multiple PDFs and splits chunks along section boundaries. Output is a NetworkX graph linking sections, tables, figures, cross-references and defined terms, with an interactive 3D/2D viewer. It uses LangChain LCEL, with PyMuPDF as the default parser and Docling as an option.
A local-first requirement & document graph: import Markdown/TXT/JSON/CSV into a per-project SQLite database and query requirements, dependencies and impact through MCP (Codex) plus a persistent local web UI.
To associate your repository with the document-graph topic, visit your repo's landing page and select "manage topics."