assafelovic /gpt-researcher
An autonomous agent that conducts deep research on any data using any LLM providers
★ 30,708⑂ 4,227👁 168⑃ 1,301● 762▤ 75
+19
Daily
+241
Weekly
+663
Monthly
30,708
stars
4,227
forks
168
watchers
1,301
prs
762
issues
75
releases
README
!GitHub Release
English | 中文 | 日本語 | 한국어 | Русский
🔎 GPT Researcher
GPT Researcher the first open deep research agent designed for both web and local research on any given task.
The agent produces detailed, factual, and unbiased research reports with citations. GPT Researcher provides a full suite of customization options to create tailor made and domain specific research agents. GPT Researcher goes well beyond RAG: planner and execution agents break each task into research questions, decision models keep only the context that is actually useful for answering them, multi-agent workflows coordinate specialized agents from planning to publication, and recursive deep research explores a topic in both breadth and depth. Research runs in parallel, which keeps it fast, stable and reliable, and reduces misinformation.
Our mission is to empower individuals and organizations with accurate, unbiased, and factual information through AI.
Why GPT Researcher?
- Objective conclusions for manual research can take weeks, requiring vast resources and time.
- LLMs trained on outdated information can hallucinate, becoming irrelevant for current research tasks.
- Current LLMs have token limitations, insufficient for generating long research reports.
- Limited web sources in existing services lead to misinformation and shallow results.
- Selective web sources can introduce bias into research tasks.
Demo
Install as Claude Skill
Extend Claude's deep research capabilities by installing GPT Researcher as a Claude Skill:
bash
npx skills add assafelovic/gpt-researcher
Once installed, Claude can leverage GPT Researcher's deep research capabilities directly within your conversations.
Architecture
The core idea is to utilize 'planner' and 'execution' agents. The planner generates research questions, while the execution agents gather relevant information. The publisher then aggregates all findings into a comprehensive report.
Steps:
Create a task-specific agent based on a research query.
Generate questions that collectively form an objective opinion on the task.
Use a crawler agent for gathering information for each question.
Summarize and source-track each resource.
Filter and aggregate summaries into a final research report.
Tutorials
- How it Works
- How to Install
- Live Demo
Features
- 📝 Generate detailed research reports using web and local documents.
- 🖼️ Smart image scraping and filtering for reports.
- 🍌 AI-generated inline images using Google Gemini (Nano Banana) for visual illustrations.
- 📜 Generate detailed reports exceeding 2,000 words.
- 🌐 Aggregate over 20 sources for objective conclusions.
- 🎯 Smart context filtering with Jev, with no API key or embeddings required by default.
- 🖥️ Frontend available in lightweight (HTML/CSS/JS) and production-ready (NextJS + Tailwind) versions.
- 🔍 JavaScript-enabled web scraping.
- 📂 Maintains memory and context throughout research.
- 📄 Export reports to PDF, WordTop issues
#1764Research is getting stucked every other time while giving no meaningful logs / ui hints indicating what it was attempting to do6 comments · 2026-05-07#1582GPL License Alternative PDF Scraper Implementation2 comments · 2026-01-05#2277PyMuPDFScraper returns empty titles for PDFs without document metadata0 comments · 2026-11-09#2295security: SSRF guard from #1818 (0.16.1+/0.17.0) is bypassable via HTTP redirect - internal content still reaches the report0 comments · 2026-11-12
Discussions
#2304Exploring Moli with GPT Researcher0 comments · 2026-11-13#2297Hi from a new contributor 👋0 comments · 2026-11-12#2288An invitation from Civitas — a city for AI agents (founding citizen № 003 reserved for GPT Researcher)0 comments · 2026-11-12#2244GPT Researcher on an OpenAI-compatible endpoint: OPENAI_BASE_URL, the three LLM slots, embeddings, and why custom model IDs can trip temperature/reasoning handling0 comments · 2026-10-30#2241Before researching podcast notes: keep the interview corpus separate from corroboration0 comments · 2026-10-29#2235Provider-portability notes: 21 places where OpenAI/Gemini specifics leak past the LLM abstraction1 comments · 2026-10-28#2223Design partner invite: independent judge for GPT Researcher agents0 comments · 2026-10-21#2211Showcase: ATS Zero-Browser Gateway — Fast protocol scraper & MCP server for career & talent research0 comments · 2026-10-15
Latest release
New paper UI, detailed reports in 3:27, and 53 community fixes v3.8.0 2026-11-09A new UI, and detailed reports in under 4 minutes The Next.js app has a new paper design that matches gptr.dev. Settings live in one modal with tabs for Research, MCP, Documents and Backend; a run shows a plain-language feed, and the sources margin slides in once writing starts. The previous UI is still available at /classic. Detailed reports now finish in 3:27 instead of 9:11. Sections are researched in parallel, and the default writer is now gpt-6-sol, which starts writing within seconds instead of reasoning for close to a minute per section first. Measurements and the model comparis
相关报道 (4)
News & coverage (8)
opentools.aiGitHub Copilot CLI vs GPT Researcher2026-05-05opentools.aiAutoGPT (Hugging Face) vs GPT Researcher2026-05-04MediumThe Modern OSINT Stack: From Search Tools to Agentic Intelligence Systems2026-07-10The RegisterAnthropic won't own MCP 'design flaw' putting 200K servers at risk, researchers say2026-04-16OX SecurityMCP Supply Chain Advisory: RCE Vulnerabilities Across the AI Ecosystem2026-04-15AutoGPT20 AI Agent Examples in 20252025-03-13R&D WorldSTORM AI can write basic Wikipedia-esque articles in minutes2024-07-24MarkTechPost‘GPT Researcher’: An Autonomous AI Agent Designed for Comprehensive Online Research on a Variety of Tasks2024-06-19