Deep Research
Agents that research a topic across many sources. As of 2026-11-21, 236 indexed products match this subcategory (79 with full profiles); 119 were published in the last 90 days, against 56 in the 90 days before, slower than the catalogue as a whole. By GitHub stars, deer-flow leads, holding 35% of the stars of the 159 products here with their own repository. The most common form is MCP servers (82% of indexed products).
SubcategorySegment label: Thin DER
Segment statistics
Whole catalogue (all 236 indexed products, not only profiled ones). Calculated 2026-11-21.
Thin. Among the smallest 30% of nodes at its level, and launches are not outpacing the catalogue. Size is ranked against the other subcategory nodes; momentum compares this node’s launches in the last 90 days with the 90 before, relative to the whole catalogue’s ratio (score 62 or more = faster, 38 or less = slower; below 5 dated launches it is not measured and the label uses size alone).
Concentration. The Herfindahl–Hirschman index (HHI) adds up each project’s squared share of stars. At 0.174, attention is spread as if across about 6 equally popular projects. Using the common competition-policy thresholds (0.15 and 0.25), that is moderately concentrated. Stars measure attention, not usage or quality.
Which product forms serve it
The same need is met by MCP servers, agent skills, plugins, extensions and frameworks. This counts each form across the whole catalogue.
| Product type | Indexed | Share | With a profile |
|---|---|---|---|
| MCP servers | 193 | 81.8% | 37 |
| Agent skills | 19 | 8.1% | 19 |
| Frameworks | 10 | 4.2% | 10 |
| Plugins | 7 | 3.0% | 6 |
| Infrastructure | 4 | 1.7% | 4 |
| Agent projects | 2 | 0.85% | 2 |
| Platforms | 1 | 0.42% | 1 |
2 capabilities
Indexed = whole catalogue; profiled = products with their own page. A product can sit in several, so the rows do not add up to the total.
| Capability | Indexed | Profiled | New, 90 days | Label |
|---|---|---|---|---|
| Automated Research ReportsProducing cited research reports automatically. | 135 | 54 | 56 | Lagging the catalogue |
| Fact-checkingVerifying claims against sources. | 36 | 12 | 17 | Thin |
Top 40 products by GitHub stars
Search the directoryProfiled products only (79 of 236), ranked by the stars of their own repository. Products that live in a subfolder of a larger repository (common for skills) show that whole repository's stars and are ranked after them; products without a repository come last. Stars are popularity, not quality or installs.
| # | Product | Type | Publisher | Capabilities | GitHub stars | Last push | Licence | Save and compare |
|---|---|---|---|---|---|---|---|---|
| 1 | deer-flowAn open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours. | Agent project | bytedance | 86k | today | MIT | ||
| 2 | BettaFish微舆:人人可用的多Agent舆情分析助手,打破信息茧房,还原舆情原貌,预测未来走向,辅助决策!从0实现,不依赖任何框架。 | Framework | 666ghj | 43k | 1mo ago | GPL-2.0 | ||
| 3 | gpt-researcherAn autonomous agent that conducts deep research on any data using any LLM providers | MCP server | assafelovic | 31k | 9d ago | Apache-2.0 | ||
| 4 | AI-Research-SKILLsComprehensive open-source library of AI research and engineering skills for any AI model. Package the skills and your claude code/codex/gemini agent will be an AI research agent with full horsepower. Maintained by Orchestra Research. | Agent skill | Orchestra-Research | 15k | 1mo ago | MIT | ||
| 5 | MiroThinkerMiroThinker is a deep research agent optimized for complex research and prediction tasks. Our latest models, MiroThinker-1.7, achieves 74.0 and 75.3 on the BrowseComp and BrowseComp Zh, respectively. | Framework | MiroMindAI | 8.5k | 4mo ago | Apache-2.0 | ||
| 6 | deep-searcherOpen Source Deep Research Alternative to Reason and Search on Private Data. Written in Python. | Infrastructure | zilliztech | 8.4k | 2mo ago | Apache-2.0 | ||
| 7 | claude-scholarSemi-automated research assistant for academic research and software development. Supports Claude Code, Codex CLI, Kimi Code CLI, and OpenCode across ideation, coding, experiments, writing, and publication. | Agent project | Galaxy-Dawn | 6.1k | 1mo ago | MIT | ||
| 8 | paper-orchestraPaperOrchestra is a multi-agent framework for automated AI research paper writing. | Framework | google-research | 5.6k | 6d ago | Apache-2.0 | ||
| 9 | exaExa AI web search, deep research, and content extraction. Provides MCP tools and research skills for comprehensive web search, people discovery, company research, academic papers, and more. | Plugin | exa-labs | 5.3k | today | MIT | ||
| 10 | hyperresearchConvert Claude Code or Codex into the most intelligent Deep Research Agent. Collect, search, and synthesize web research into a persistent, searchable wiki that builds on itself. Hosted API + MCP: hyperresearch.ai | MCP server | jordan-gibbs | 5.0k | 4d ago | MIT | ||
| 11 | deep-researchUse any LLMs (Large Language Models) for Deep Research. Support SSE API and MCP server. | MCP server | u14app | 4.7k | 5mo ago | MIT | ||
| 12 | pipeshub-aiThe open-source context layer for AI agents and enterprise AI search. Permission-aware search and cited answers across 30+ sources, including Slack, Drive, Jira, GitHub and Microsoft 365, for people and agents through MCP, SDKs and built-in agents. Self-hosted or Cloud. | MCP server | pipeshub-ai | 4.1k | today | Apache-2.0 | ||
| 13 | MiroFlow🏆 Top-1 on 5+ benchmarks | Web UI | Supports MiroThinker, Claude, Kimi, OpenAI | Framework | MiroMindAI | 3.2k | 4mo ago | Apache-2.0 | ||
| 14 | Deep-Research-skillsStructured deep research skill for Claude Code/Open Code/Codex with human-in-the-loop control | Agent skill | Weizhena | 2.6k | 3mo ago | MIT | ||
| 15 | last30days-skill-cn给 AI Agent 的中文互联网研究 Skill:检索 8 大平台最近 30 天的讨论,追踪跨平台热点,生成附原始来源的报告,并提供每日全网热榜。支持 Claude Code、Codex、OpenClaw。 | Agent skill | Jesseovo | 2.2k | 1d ago | — | ||
| 16 | paperthinThe skillbook for recursive self-improvement (RSI). NVIDIA's latest proof, unpacked with timeless engineering. Distilled from the OpenAI harness engineering hackathon winner's zero-touch method. | Agent skill | LilMGenius | 1.5k | 6d ago | MIT | ||
| 17 | NanoResearch🦞+🔬 NanoResearch: The Autonomous AI Research Assistant | Agent skill | OpenRaiser | 1.3k | 1mo ago | MIT | ||
| 18 | enterprise-deep-researchSalesforce Enterprise Deep Research | Framework | SalesforceAIResearch | 1.2k | 25d ago | Apache-2.0 | ||
| 19 | pydantic-deepagentsOpen-source, self-hosted Claude Code - a terminal AI assistant and the Python framework behind it. Tool-calling, sandboxed execution, multi-agent teams, skills, checkpoints, unlimited context - on Pydantic AI, any model. | Framework | vstorm-co | 1.1k | 22d ago | MIT | ||
| 20 | codex-slides🎨 Open-source AI slide studio inside Codex: image-native decks, every slide a full visual canvas. ⚡ 10+ high-quality slides in ~4–5 minutes — Fast mode renders every page in parallel. 🔍 Watch the whole chain live: research → outline → style → render → edit → present → export PDF/PPTX. 🖥️ Browser-first · zero API keys · durable projects. | Agent skill | nexu-io | 987 | 4mo ago | MIT | ||
| 21 | openagentWhat if OpenAI Deep Research and Dify were one platform? OpenAgent — harness architecture for rapidly building vertical AI agents, with deep reasoning loops, visual workflows, RAG, and A2A delegation. | Infrastructure | Haohao-end | 756 | 1d ago | MIT | ||
| 22 | ClaudeSkills13 curated Agent Skills for research, product decisions, decks, publishing, audits, and more — portable across skills-compatible agents. | Agent skill | staruhub | 741 | 3mo ago | MIT | ||
| 23 | claudemyClaudemy: an open-source alternative to Udemy and Coursera. Give Claude a topic, it researches the web and builds a course of 2-minute audio lessons with cited sources. A free Claude Code skill. | Plugin | assafkip | 713 | 23d ago | MIT | ||
| 24 | deerflow-book本书围绕 DeerFlow 2.0,从理论到源码,系统讲解如何进行二次开发。(精校) | Framework | hawkli-1994 | 595 | 2d ago | MIT | ||
| 25 | FinSight-AIAI equity research agent with resilient workflows, evidence-grounded RAG, versioned reports, and automated quality evaluation. | Infrastructure | juanjuandog | 562 | 2mo ago | MIT | ||
| 26 | CoexistAICoexistAI is a modular, developer-friendly research assistant framework . It enables you to build, search, summarize, and automate research workflows using LLMs, web search, Reddit, YouTube, and mapping tools—all with simple MCP tool calls or API calls or Python functions. | MCP server | SPThole | 531 | 8mo ago | — | ||
| 27 | designing-real-world-ai-agents-workshopHands-on workshop: Build a multi-agent AI system from scratch — Deep Research Agent + Writing Workflow served as MCP servers. Includes code, slides, and video | Framework | iusztinpaul | 523 | 5mo ago | MIT | ||
| 28 | huashu-report机构级研究报告 Agent Skill · 规范从 2026 年顶级机构报告实物反向提炼,42 份采集/41 份进量化基线(Stanford/McKinsey/BCG/OpenAI/PwC/World Bank 等)· 六种报告原型 + 8 种图表模式 + 可复用生产流水线 · 跨 agent 通用 | Agent skill | alchaincyf | 503 | 6d ago | MIT | ||
| 29 | dsh-desktopDeepSeek Harness Desktop App: a local AI desktop workspace for DSH Sessions, projects, files, web research, plugins, and Office artifacts. | MCP server | vibeinging | 497 | today | MIT | ||
| 30 | atlas-mcp-serverA Model Context Protocol (MCP) server for ATLAS, a Neo4j-powered task management system for LLM Agents - implementing a three-tier architecture (Projects, Tasks, Knowledge) to manage complex workflows. Now with Deep Research. | MCP server | cyanheads | 480 | 1y ago | Apache-2.0 | ||
| 31 | ai-skills24 cross-platform agent skills for Claude Code, Cursor, Codex & Gemini CLI — databases, messaging, research, TTS, DevOps, and Google Workspace | Agent skill | sanjay3290 | 457 | 18d ago | Apache-2.0 | ||
| 32 | color-spaceConversions across 221 color spaces, tested against cited sources: CSS strings, gamut, LUTs | MCP server | colorjs | 397 | 9d ago | CC0-1.0 | ||
| 33 | blockrun-mcpWeb search, deep research, prediction markets & crypto data for AI agents. Pay per call via x402. | MCP server | BlockRunAI | 390 | 3d ago | MIT | ||
| 34 | gptr-mcpMCP server for enabling LLM applications to perform deep research via the MCP protocol | MCP server | assafelovic | 380 | 1y ago | MIT | ||
| 35 | humanizer-skillSee why a draft sounds like AI and rewrite it in one of five voices: 55 named patterns, a 0 to 100 score, a fact check. | Plugin | Aboudjem | 325 | 3d ago | MIT | ||
| 36 | agent-systems-handbookAgent Systems Handbook by Prompthon — A practical AI agents handbook covering agent systems, agentic workflows, LangGraph, MCP/A2A, context engineering, agent memory, evaluation, observability, and multi-agent architecture. Current trend focus: Gemini Interactions API and managed agents, emerging agent runtimes, and production AI workflow patterns. | Framework | Prompthon-IO | 315 | 1mo ago | — | ||
| 37 | research-toolkitResearch methods, workflow tools, and independent review for AI agents, packaged as a Skill, plugin, and local MCP server. | MCP server | rrrrrredy | 314 | 1d ago | MIT | ||
| 38 | enrichableOpen-source AI spreadsheet for data enrichment. Build lead lists from a prompt, research every row with live web data, and find work emails. Powered by Context.dev. | MCP server | context-dot-dev | 285 | 2d ago | MIT | ||
| 39 | argoMulti-engine search and deep research for AI agents across 300+ sources, 18 languages, and MCP / CLI. | MCP server | taxueseek | 284 | 3d ago | Apache-2.0 | ||
| 40 | research-starter-skills面向科研新手的 Codex 科研协作技能集:选题、文献检索与阅读、论文写作、自然润色、审稿返修和可编辑学术汇报。结合 Research-Starter-Kit 原创教程,用 1 个总入口与 6 个专业模块推进研究。 Research training, writing & editable presentations for Codex. | Agent skill | YCC-Lover | 263 | 21d ago | — |
Most common integrations
Third-party services named by products in this node, whole catalogue.
| Integration | Products | Share |
|---|---|---|
| GitHub | 3 | 1.3% |
| 3 | 1.3% | |
| YouTube | 2 | 0.8% |
| Ethereum | 1 | 0.4% |
| Google Workspace | 1 | 0.4% |
| Jira | 1 | 0.4% |
| Neo4j | 1 | 0.4% |
| Salesforce | 1 | 0.4% |
| Slack | 1 | 0.4% |
Launches per month
New Deep Research products per monthSRC
Whole catalogue, by publication month, 2024-12 to 2026-11. The last month is partial; undated products are excluded.
Show the data as a table
| Products published (products) | |
|---|---|
| 2024-12 | 1 |
| 2025-02 | 2 |
| 2025-03 | 2 |
| 2025-05 | 1 |
| 2025-06 | 2 |
| 2025-08 | 3 |
| 2025-09 | 2 |
| 2025-10 | 2 |
| 2025-11 | 2 |
| 2025-12 | 2 |
| 2026-01 | 4 |
| 2026-02 | 2 |
| 2026-03 | 8 |
| 2026-04 | 8 |
| 2026-05 | 11 |
| 2026-06 | 10 |
| 2026-07 | 20 |
| 2026-08 | 32 |
| 2026-09 | 37 |
| 2026-10 | 49 |
| 2026-11 | 25 |
Related opportunities
Opportunity radarScored gaps for this node or anything beneath it. Scores rank hypotheses; they do not predict demand.
| Opportunity | Competition | Confidence | Score |
|---|---|---|---|
| Coverage gap: Fact-checkingFact-checking | High | Low | 54.2 |
Limits of these numbers
- Products are placed here by whole-word keyword rules on their name, description and repository topics, and inherit every ancestor of a matched node. Products that do this without naming it are missed; ambiguous words cause false matches. Methodology
- About half of the catalogue matches no node at all, so absolute counts understate every category.
- Launch counts use the publication date each source reports; for auto-discovered repositories that is the repository’s creation date.
- Stars and concentration cover only products with their own repository (not a subfolder of a larger one). They measure attention, not adoption.
Page data last updated .