by lotchuazzz-crypto · MCP Server · ★ 285
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is papergraph-mcp safe to install? View the security audit →
PaperGraph MCP PaperGraph turns local or arXiv LaTeX papers into theorem dependency graphs that AI agents can query through MCP. PaperGraph v0.4.4 hardens first-use analysis by making agents verify the running version, validate raw arXiv requests, and report dependency boundaries before interpreting sparse graphs. PaperGraph v0.4.0 introduced the persistent, cross-paper SQLite workspace: retain a small literature collection, search theorem text, follow theorem dependencies, and inspect citation evidence without asking an agent to re-read every source paper. Why PaperGraph? Single-paper tools expose theorem-like environments, labels, and relationships. A workspace keeps many independently imported papers together. It is deliberately evidence-first: every citation result identifies the source paper, bibliography key, LaTeX command, source file, and resolution status. This is useful for grounded reading and review; it is not semantic theorem matching or a claim that two similarly worded
uvx --from git+https://github.com/lotchuazzz-crypto/papergraph-mcp.git@v0.4.4 papergraph-mcp --versionTaken from the project's README on GitHub. Check the security audit before running it.
No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
| Stars | 285 |
| Forks | 4 |
| Language | Python |
| Category | MCP Server |
| License | MIT |
| Quality Score | 71.5173946339524/100 |
| Open Issues | 1 |
| Last Updated | 2026-10-15 |
| Created | 2026-09-02 |
| Platforms | mcp, python |
| Est. Tokens | ~16k |
These tools work well together with papergraph-mcp for enhanced workflows:
Looking for a papergraph-mcp alternative? If you're comparing papergraph-mcp with other mcp server tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
CLI server for Zotero 7/8/9 — let AI manage your library. 70+ CLI commands for search, import, PDF, BibTeX, no
An IOS Simulator Skill for ClaudeCode. Use it to optimise Claude's ability to build, run and interact with you
Stock market MCP server for Yahoo Finance data, technical analysis, portfolio tracking, and Python backtesting
Local codebase intelligence CLI + MCP server for AI coding agents: SQLite code graph, 28 languages, 287 comman
Structural code intelligence for AI agents — semantic search, knowledge graphs, and a built-in MCP server in o
Human-like memory for AI agents — semantic, episodic & procedural. Experience-driven procedures that learn fro
Explore other popular mcp server tools:
papergraph-mcp is PaperGraph 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.. It is categorized as a MCP Server with 285 GitHub stars.
papergraph-mcp is primarily written in Python. It covers topics such as ai-agents, arxiv, knowledge-graph.
The README on GitHub (github.com/lotchuazzz-crypto/papergraph-mcp) gives: uvx --from git+https://github.com/lotchuazzz-crypto/papergraph-mcp.git@v0.4.4 papergraph-mcp --version. Usage details are in the same README.
papergraph-mcp is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to papergraph-mcp on Agent Skills Hub include cli-anything-zotero, ios-simulator-skill, maverick-mcp. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: