Vibe Coding Discoverpapergraph-mcp
View on GitHubPaperGraph MCP turns math papers into evidence-grounded reading maps for AI agents: extract results, trace proof evidence, plan reading order, and review external dependencies without guessing.
PaperGraph MCP is a stdio MCP server that turns arXiv papers, local LaTeX and PDFs into an evidence-grounded theorem dependency workspace. Agents can build Paper Maps, trace proof and citation evidence, plan reading order, resolve blocked references, and export Markdown reading reports without guessing dependencies.
Use Cases
Let an AI agent load an arXiv math paper and build a theorem-centric Paper MapTrace proof evidence and source spans for a target theoremGenerate a reading queue from local proof dependenciesExport Markdown Reading Reports for handoff or GitExport cross-paper reading plans across a few related papersResolve blocked references via Crossref, OpenAlex and arXiv metadata searchPlan and review external paper import plans before fetchingPersist reading sessions, checkpoints and notes in a local SQLite workspaceDiagnose sparse evidence extraction with Evidence TriageParse local LaTeX projects or born-digital PDFs instead of arXiv
Built With
- Language
- Python
- Frameworks
- MCP Python SDK · FastMCP · PyMuPDF · pybtex · httpx · pytest · uv · hatchling
Tags
mcp · mcp-server · arxiv · latex · mathematics · knowledge-graph · theorem-dependency-graph · evidence-extraction · research-tools · pdf · reading-workflow · python · stdio · ai-agents · skills · citations