paper-fetch-skill

Based on the current evidence, this is worth watching, but it is better judged as a focused open-source component than as a proven, mature academic retrieval platform. The only strong source here is the official GitHub repository copy, which supports its stated core function: converting DOI, URL, or title inputs into agent-ready structured markdown/full text. The 222 GitHub stars and 27 forks show attention and interest, but they do not by themselves prove retrieval reliability, coverage, or long-term maintenance quality.

In practice, it looks more like a paper ingestion and formatting layer for AI workflows than a discovery engine like Semantic Scholar or Google Scholar, and not a reference manager like Zotero. A better analogy is a paper-ingestion skill for LLM agents, scripts, and research pipelines: resolve a paper entry point, then normalize it into text that is easier to summarize, cite, or feed into downstream analysis. That makes the use case fairly concrete for RAG, literature summarization, and research automation.

On cost and adoption, the evidence only supports that it is open source and offers CLI plus MCP integration.