Moonshot AI's assistant, known for very long context windows and strong agentic coding and research workflows.
Research
Gathering and synthesising information from literature, papers, datasets and the open web.
24 apps, 7 skills and 12 MCP servers tagged Research.
Apps
AI for systematic literature review — search 125 million papers, extract data into tables, and screen at scale.
Microsoft's open-source optimizer that trains an agent's SKILL.md like a model parameter — epochs, learning-rate budgets and a held-out validation gate, with frozen weights.
Web data extraction APIs for AI agents — 75+ ready-made scrapers for LinkedIn, Amazon, Google Maps, Reddit and X, priced per result.
Deep-research AI agent that builds cited reports or structured datasets, scaling depth to your budget.
Synthetic user research: interview AI personas about your product, messaging or landing page and get a structured report back in about 30 minutes.
Audit-grade fact-checking API for AI products — eight models across five stages, adversarial debate, and a full citation trail behind every verdict.
Personal intelligence agent that tracks the topics you name across news, filings, forums and podcasts, folds duplicate coverage into one card, and briefs you by feed, push, email, RSS or audio.
Build systematic investment strategies with AI, validate them on out-of-sample and live paper data, then let an MCP-connected agent mirror them in your own brokerage account.
An approval-first agent workspace for solo founders: planning, research, social and coding agents propose one next move at a time and only act once you say yes.
AI investment agents that research markets, monitor portfolios and place orders on your instruction, with every action logged for review.
An AI research assistant over 280 million papers — chat with PDFs, run literature reviews, and decode dense text.
A clinical answer engine for doctors, grounded in peer-reviewed medical literature with citations on every claim.
Enter one paper and get a visual graph of the work around it — prior art, derivatives, and adjacent research.
Collaborative legal AI for review, research, and drafting — built to work inside the documents lawyers already use.
Domain-specific AI for elite law firms and corporate legal teams — research, drafting, and diligence at scale.
A deep-search research assistant that reads full papers, not abstracts, to find work others miss.
Citation-graph exploration for literature reviews — start from one paper and map the field around it visually.
An all-in-one agentic workspace that turns a single prompt into slides, documents, sheets, images, video, or a working page.
Scite Assistant answers questions with cited science.
Mathpix converts math and science from images to LaTeX.
Wolfram Alpha is a computational knowledge engine.
Tana is a networked workspace with AI commands.
One place for all of your ideas, notes, highlights, and writing.
Skills
Tencent's skill-evolution harness: it rewrites a whole skill folder — SKILL.md, scripts and references together — and lands every decision as a real Git issue, PR and wiki entry you can review.
Multi-source academic literature search as an agent skill — OpenAlex, Semantic Scholar, CrossRef, PubMed and arXiv at four depths, from a 5-minute scan to a PRISMA-S logged systematic review.
A four-skill bundle that runs the academic pipeline end to end — deep research, paper writing, peer review and revision — as multi-agent workflows inside Claude Code.
Turns a plain-language weather question into a working Earth2Studio inference script — picks the AI forecast model, a compatible data source, an IO backend, and the step count.
Skill: Brave Web Search
by Brave
Brave's official skill for the Brave Search API — ranked web results with snippets and thumbnails, freshness and SafeSearch filters, search operators, pagination, and Goggles for custom ranking.
Skill: Brave Search CLI (bx)
by Brave
Brave's official search skill: one CLI that searches, scrapes and extracts in a single call, returning token-budgeted web content instead of a list of links.
Skill: Design Harness
by tigerless-labs
A decision board for contested calls: feed your agent papers and half-formed ideas, and it files them into a traceable sources → ideas → output structure you can defend, with a visual canvas over the markdown.
MCP servers
Vulnerability triage for AI agents: 28 security tools across 24 free intelligence sources — NVD, EPSS, CISA KEV, MITRE ATT&CK, Shodan and VirusTotal — behind a single triage_cve call.

MCP: Altmetric MCP Server
by Altmetric
Altmetric's official MCP server for research attention data — news, policy documents, patents and social mentions of a paper — over a hosted OAuth endpoint or locally with your own API keys.
MCP: Patent & Literature Search
by Patsnap
Patsnap's MCP server for searching global patents across WIPO, EPO, USPTO, CNIPA, JPO and KIPO alongside scientific literature, returning records as AI-readable Markdown.
MCP: Hugging Face
by Hugging Face
Official Hugging Face MCP server: search models, datasets, Spaces and papers on the Hub, read repository files, and call Gradio apps as agent tools.
MCP: BioMCP
by GenomOncology
Biomedical research MCP server reaching about 30 trusted sources — PubMed, ClinicalTrials.gov, ClinVar, OncoKB and Reactome — through one command grammar.
MCP: wigolo
by KnockOutEZ
Local-first web MCP server for AI agents — search, fetch, crawl, extract and deep research with no API keys, no cloud round-trip and no per-query bill.
MCP: PubMed
by Andy Brandt
Search PubMed's biomedical literature and read abstracts and open-access full text from an AI client.
MCP: arXiv
by Joseph Blazick
Search arXiv, download papers, and read their full text from an AI client — research without the PDF shuffling.
MCP: YouTube Data
by icraft2170
Search YouTube, read video transcripts, and pull channel statistics from an AI client.

MCP: Pipeworx
by Pipeworx
Open MCP gateway giving agents live, citable data from 1,300+ verified sources like SEC, FDA, USPTO, and FRED.
MCP: Perplexity Ask
by Perplexity
Ask Perplexity's Sonar API real-time, web-grounded questions and get cited answers inside your agent.
MCP: Sequential Thinking
by Model Context Protocol
A tool for dynamic, reflective problem-solving through structured, revisable sequences of thoughts.
Related tags
Tags that appear alongside this one, ranked by how often.