GPT Researcher
Autonomous research agent that produces sourced, structured reports
View on GitHubWhat it solves
Research that takes hours of manual reading: the search, scrape, and synthesize loop runs automatically and returns a structured, sourced report instead of a list of links.
Overview
GPT Researcher takes a research question, runs parallel web searches, scrapes and reads sources, synthesizes findings, and produces a cited, structured report, without human guidance between steps. The agent iteratively refines its queries based on what it finds, working like a research assistant that reads dozens of pages before writing. Outputs Markdown or PDF; supports custom data sources alongside open web search.
Key facts
- Language
- Python
- License
- Apache 2.0
- Maturity
- Stable
- Maintainer
- assafelovic
- Reviewed
- 2026-06-11
Where to start
The architecture doc explains the Planner–Researcher–Publisher agent split, knowing who owns which step clarifies what to customize before you write any configuration.
docs/docs/gpt-researcher/getting-started/getting-started.mdTechnologies
More in this domain
LangGraph
Stateful multi-agent orchestration for LLM applications
The state management problem in multi-agent systems: how you build agents that pass context across turns without losing the execution thread.
AutoGen
Multi-agent conversation framework from Microsoft Research
How to coordinate multiple specialized agents toward a shared goal without writing low-level orchestration logic by hand.
CrewAI
Role-based multi-agent orchestration for collaborative task execution
The boilerplate problem in multi-agent setup: defining roles, delegation rules, and task routing without writing a custom orchestration graph.