Describe a topic. Three specialist agents hand the work to each other and deliver a structured, export-ready report, while you watch every step live.
MISSION CONTROLEach agent has one job, a clear hand-off and a visible status, so you always know what the crew is doing.
A Researcher gathers facts, an Analyst weighs them and an Editor writes the final report.
Watch progress on the reactor, agent cards and a timestamped mission log as the work moves along.
Pick a short, standard or deep report and choose a professional, friendly, academic, persuasive or simple tone.
Every finished report is saved automatically so you can reopen, review or delete it later.
Copy the text, download clean Markdown, or print the report to PDF from your browser.
CrewAI for the agents, Groq for the model and FastAPI for the backend. Swap the model in one line.
Type your topic and choose the length and voice you want.
Researcher, Analyst and Editor pass the work along while you follow each step live.
Get a structured Markdown report with summary, findings, analysis and recommendations.
Get an organised starting point for assignments, presentations and exam topics.
Scan a market, technology or idea before a meeting or a decision.
Draft a structured briefing quickly, then refine it with your own sources.
By default an open GPT-OSS model served through Groq. You can change the model name in the .env file.
The agents are told not to invent facts or references, but AI can still make mistakes. Always verify important claims against trusted sources.
On your own machine, in the reports folder next to the app. Nothing is uploaded anywhere except the prompts sent to the model provider.
Yes. Add your Groq API key to the .env file before starting the server.
Give it a topic and see the report come together in minutes.
Launch crew