In simple English
Company research takes hours because details hide across websites, news pages, and social profiles. Mira automates this work by sending multiple artificial intelligence agents to search for you. The agents visit company sites, read LinkedIn pages, and run web searches to build one structured profile. When the agents find all needed facts with high confidence, the system stops early to save your money. You can use the core library in your own code, or run the included Next.js web application.
What it is
Mira is an open-source, multi-agent research toolkit built in TypeScript that gathers, verifies, and scores company information. It combines automated web scraping, search queries, and confidence scoring into a single package.
What it is good for
Solo founders and sales teams who need to gather background information on hundreds of prospective client companies without manual searching.
- Stops searching early once confidence scores are met, saving API costs.
- Assigns a confidence score and source reference to every gathered fact.
- Includes a ready-to-use web interface with workspace management.
- Requires paid external API keys for OpenAI and ScrapingBee to function.
- Setting up the web interface requires configuring a separate database service.
- Node.js version 18 or later is required to run the code.
manual web research and copy-pasting company data into spreadsheets
Supabase
- langchain-ai/langgraph44k stars
A framework for building stateful multi-agent applications with code rather than a pre-packaged research workflow.
- apache/airflow47k stars
Runs chains of scheduled jobs written in Python, and keeps a record of every run.
- cybertec-postgresql/pg_timetable1.4k stars
A job scheduler for PostgreSQL whose jobs and schedules live in the database itself.
git clone https://github.com/dimimikadze/mira.git cd mira npm install