celesteblackandwhite925/paper-distill-mcp

This is a thin wrapper around existing academic paper APIs (arXiv, Semantic Scholar, etc.) using the Model Context Protocol as the transport layer. The MCP specification itself is defined by Anthropic; this project simply implements an MCP server that wraps standard paper search functionality. **Defensibility Gaps:** - Zero stars and 1 fork after 17 days indicates no adoption or community validation - Velocity is 0 (no recent activity), suggesting the project is stalled or abandoned - The core capability (paper search) is commodity functionality available through direct API access or numerous existing tools (Google Scholar, Semantic Scholar, Papers.co, Elicit, etc.) - MCP is a new protocol with limited adoption; implementing MCP servers for niche domains is not yet defensible - No apparent novel algorithm, data source, or curation strategy that differentiates from existing paper discovery tools **Platform Domination Risk (HIGH):** - OpenAI, Anthropic, and other AI labs are actively building paper discovery into their platforms (ChatGPT's research browsing, Claude's document analysis) - Anthropic designed MCP specifically to enable integrations like this; they could trivially bundle paper search as a native MCP capability - Major cloud providers (AWS, Google Cloud) are building research corpus indexing into their services - Paper discovery is adjacent to LLM product roadmaps and will likely be absorbed **Market Consolidation Risk (MEDIUM):** - Incumbents (Semantic Scholar, Elicit, Papers.co, Zotero) already provide paper search with curation layers - These incumbents have:funding, existing user bases, and established data partnerships with arXiv, PubMed, etc. - However, none have explicitly prioritized MCP integration yet, leaving a small window before they do - Acquisition is possible if the project gained traction (unlikely at current velocity) **Displacement Horizon (6 MONTHS):** - Anthropic or a competitor could launch native paper search MCP in weeks - Existing paper discovery incumbents will integrate MCP support once adoption grows - The project has no defensible lead; displacement is imminent once this niche becomes visible **Why Score 2, Not 1:** - The project is functionally deployable (not a tutorial) - MCP is a legitimate protocol gaining traction in AI applications - If the project added domain-specific curation (domain-specific paper ranking, citation network analysis, personalization) it could climb to 3-4 - Current state is purely commodity aggregation with no defensibility **Recommendation:** This project has no moat and faces immediate displacement risk. To become defensible (5+), it would need: (1) proprietary paper ranking or filtering, (2) deep integration with research workflow tools, (3) niche domain expertise (e.g., a specialized corpus for biomedical vs. ML), or (4) community adoption that creates switching costs. Current trajectory suggests it will be obviated by platform-native solutions within 6 months.