GPT Researcher
Autonomous AI research agent that searches the web and produces detailed, cited research reports in minutes.
GPT Researcher
GPT Researcher is an autonomous agent designed to perform comprehensive online research on any topic and produce detailed, factual reports with citations. It orchestrates multiple sub-agents — a planner and execution agents — to scrape relevant sources, extract key facts, and synthesise a structured markdown report.
Key Features
- Multi-agent orchestration: orchestrator plans subtasks, parallel agents execute web searches
- Source deduplication and credibility filtering across 20+ web sources per query
- LLM-provider agnostic: GPT-4, Claude, Gemini, or any LangChain-compatible model
- Outputs: detailed report, quick summary, resource list, or outline
- FastAPI server and React frontend included for production deployment
Quick Start
pip install gpt-researcher
export OPENAI_API_KEY=sk-...
export TAVILY_API_KEY=tvly-...
from gpt_researcher import GPTResearcher
import asyncio
async def main():
researcher = GPTResearcher(
query="Latest advances in multimodal LLMs 2025",
report_type="research_report",
)
await researcher.conduct_research()
report = await researcher.write_report()
print(report)
asyncio.run(main())
npx ai-supply add gpt-researcher-ai-research
Curated mirror of the open-source GPT Researcher (Apache-2.0). Get it from the source.
Compromise signals — malicious or tampered code (leaked secrets, backdoors, a dropped executable) — reduce the score, and known dependency CVEs carry a bounded penalty (they warrant review but never QUARANTINE — update the dependency to clear). Other dangerous-by-capability traits are risk surface, expected for some capabilities. Every finding is mapped to its OWASP control below.
Findings mapped to the OWASP Top 10 for LLM Applications (2025) and the OWASP Machine Learning Security Top 10. Expand any flagged control for the exact findings — compromise reduces the score; expected/risk-surface do not, except a known CVE, which carries a small bounded penalty (high/critical → Review).
The same gate an agent runs before installing (POST /api/v1/trust/gpt-researcher-ai-research/check). Click a policy:
Consume GPT Researcher programmatically. Authenticate with an API key or session — see Authorize an agent.
# Agents: CHECK BEFORE YOU INSTALL (no auth) — score, grade, level, capability manifest
curl https://ai-supply.store/api/v1/trust/gpt-researcher-ai-research
# Gate against your org policy (returns { pass, violations })
curl -X POST https://ai-supply.store/api/v1/trust/gpt-researcher-ai-research/check \
-H "Content-Type: application/json" \
-d '{"minGrade":"B","denyPermissions":["shell"],"denyUnknownEgress":true}'
# CLI
npx ai-supply add gpt-researcher-ai-research
# REST (install → download)
curl -X POST https://ai-supply.store/api/v1/listings/gpt-researcher-ai-research/install \
-H "Authorization: Bearer $AIM_KEY"
# MCP tool
install_listing({ "slug": "gpt-researcher-ai-research" })OpenAPI spec →Curated mirror — latest upstream source. See the repository for tagged releases.