Nice job. I put together a similar system a while back - it's just a self-contained Go binary called paper-search along with an accompanying LLM skill to facilitate search, retrieval, and downloading of relevant academic papers using OpenAlex, Semantic Scholar, and arXiv.
In my experience it's been a better solution versus just asking the LLM to directly to search the web for this kind of information via search engine tooling.
Trying this out on a valve engineering PDF extraction task — the repo claims 72% improvement on PDF extraction specifically, which is why I'm biting. Installed the MCP and the results were actually contextualized against what I'd already tried in the repo, which surprised me. Handing it to Claude Code now. Quick question: does it pull implementation details from the paper itself, or just the technique? Will report back.
Yes, we contextualize the search for technques to your work and then pull everything out : the idea itself, supporting results and all implementation details - so your coding agent has a trivial time implementing it
Nice job. I put together a similar system a while back - it's just a self-contained Go binary called paper-search along with an accompanying LLM skill to facilitate search, retrieval, and downloading of relevant academic papers using OpenAlex, Semantic Scholar, and arXiv.
In my experience it's been a better solution versus just asking the LLM to directly to search the web for this kind of information via search engine tooling.
Also just FYI the link provided in your Show HN (https://github.com/paper-lantern-ai/paper-lantern-challenges) is a 404. I think it should be:
https://github.com/paperlantern-ai/paper-lantern-challenges
yes - definitely, i have noticed the same that customized solutions for research papers works better than LLM web search.
thanks for catching the link issue
in case you can try out our solution for code agents, i'd love to hear what you think of it...
Trying this out on a valve engineering PDF extraction task — the repo claims 72% improvement on PDF extraction specifically, which is why I'm biting. Installed the MCP and the results were actually contextualized against what I'd already tried in the repo, which surprised me. Handing it to Claude Code now. Quick question: does it pull implementation details from the paper itself, or just the technique? Will report back.
Thanks for trying it out :)
Yes, we contextualize the search for technques to your work and then pull everything out : the idea itself, supporting results and all implementation details - so your coding agent has a trivial time implementing it
[dead]