Full-text semantic paper search
Use plain-language queries to search Lune's indexed corpus of top-tier computer science conference papers and find foundational, related, or recent work.
Lune Research
Lune Research connects AI agents to indexed computer science papers, citation context, full-text sources and research guidance for reviews, experiments and academic writing.
Evidence-grounded research for AI agents
Lune Research is a research retrieval service and MCP server for AI agents. It connects web AI apps and local agent clients to an indexed collection of top-tier computer science conference papers, allowing an agent to search literature, inspect sources, trace citations, and retrieve research guidance instead of relying only on model memory or open-web results.
The service is designed for workflows that move from literature discovery to experimentation and writing. Search results can include paper abstracts, authors, venue and publication metadata, citation relationships, stable source links, and full-text passages where Lune has permission to index the source. Its guidance tools cover literature reviews, evaluation design, ablation studies, research writing, peer review, and venue selection. Lune is available as a hosted web miniapp, an AI-app connector or plugin, an MCP server, and an optional command-line setup.
Research tools for AI agents
Paper and full-text inspection
Inspect paper abstracts, authors, venue and publication metadata, stable source links, and indexed source passages when full text is available.
Citation tracing
Follow references a paper cites and later indexed papers that cite it to understand research lineage and surrounding evidence.
Claim checking and evidence gathering
Check claims against the corpus, with a verbatim quote where the evidence contains one, and gather evidence across multiple searches for a defined task.
Research best-practice guidance
Retrieve curated guidance for literature reviews, evaluation design, ablations, research writing, peer review, and venue selection.
AI-client integrations
Connect the hosted MCP server or local stdio package to web AI apps, coding agents, and other MCP clients so the agent can call Lune during a research workflow.
Where Lune fits in a research workflow
Literature reviews
Ask an AI agent to identify foundational papers, related work, and possible gaps, then inspect abstracts, metadata, full text, and citation relationships before forming a literature map.
Brainstorming and novelty checks
Use the guidance library and paper corpus to examine whether an idea is already established, compare it with prior methods, and frame a research question with evidence.
Experimental design
Retrieve recommendations about expected baselines, ablations, datasets, and evaluation practices while planning experiments, then use claim checking to test specific statements against indexed sources.
Academic writing
Have an agent draft related-work or technical prose from retrieved papers while retaining source links, citation context, and checked evidence for later review.
Shared AI-assisted research
Connect Lune to a team-shared plan or an existing AI client so collaborators can use the same research service across web chat and local development workflows.
Pros and Cons
Pros
- Combines paper search, source inspection, citation tracing, claim checking, and research guidance in one agent-accessible service.
- Supports both hosted web use and integrations with common web AI apps, coding agents, and MCP clients.
- Returns source metadata and stable links to help an agent connect answers to identifiable papers.
- Provides a free tier and usage credits for additional requests, with team sharing available on paid plans.
Cons
- Coverage is focused on indexed top-tier computer science conference literature rather than the entire scholarly record.
- Full-text inspection depends on Lune having permission to index the paper; some results may provide metadata or abstracts without full text.
- Local and connector setups may require OAuth configuration or an access key, depending on the client.
FAQ
How can I use Lune Research?
Lune can be used as a hosted web miniapp, as a connector or plugin inside supported web AI apps, or through its MCP server in local clients such as Claude Code, Cursor, VS Code, Codex, OpenCode, Antigravity, Zed, and other MCP clients. An optional command-line package can help configure local connections.
What kind of research information does Lune provide?
Lune searches indexed top-tier computer science conference papers and can return abstracts, authors, venue and publication metadata, citation relationships, stable source links, and full-text passages when it has permission to index the paper. It also provides curated research guidance for methods, evaluation, writing, peer review, and venue selection.
Does Lune require an API key or access key?
The hosted MCP server uses OAuth or an access key in a Bearer authorization header. Web apps that support browser sign-in can use OAuth; apps that require a fixed credential can use an access key created from the Lune dashboard.
How is Lune Research priced?
Lune offers a free plan with 10 requests per day. The Pro plan is listed at $4.99 per month for 300 requests per day, and the Max plan at $9.99 per month for 600 requests per day. Extra usage is available as $0.99 for 300 additional requests, and credits do not expire.
What are Lune's source and full-text limitations?
Lune's indexed corpus focuses on top-tier computer science venues, including CORE A* or CCF A conferences. The site states that it excludes workshop, findings, tutorial, short, abstract, and demo papers; full text is available only where Lune has permission to index it.
Quick Facts
- Category
- AI research assistant and developer tool
- Primary users
- Students, researchers, and AI-assisted research teams
- Source focus
- Top-tier computer science conference papers, including CORE A* or CCF A venues
- Access methods
- Hosted web miniapp, web-app connector/plugin, MCP server, and optional CLI
- Hosted MCP URL
- https://mcp.luneresearch.com
- Free plan
- 10 requests per day; paid plans and non-expiring top-up credits are available
Lune Research Alternatives
CreateOS Sandbox
CreateOS Sandbox is an isolated compute environment for running code and agent workloads in Firecracker micro-VMs with private networking and SDK, CLI, or MCP control.
Hype
Hype is a web tool for finding trending YouTube topics by category, time range, and scoring mode. Spot emerging ideas and review source videos.
GitBot
GitBot is a local browser workspace for reusable coding bots powered by Claude Code, Codex, or OpenCode, with persistent threads and configurable permissions.
Codex Plugins
Codex Plugins bundle reusable skills, app integrations, and MCP servers into workflows you can install in the Codex app or use from Codex CLI.
Wallie
Wallie is an open-source AI streamer that watches your screen, hears chat, and delivers live commentary in a configurable persona. Runs locally with your own keys.
ClayHog
ClayHog is an AI search visibility platform for tracking brand presence in ChatGPT, Gemini, Perplexity, Claude, and Google AI Overviews from one dashboard.