
Web Application Security Automation Framework which recons the target for various assets to maximize the attack surface for security professionals & bug-hunters
It takes user input as a domain name and maximize the attack surface area by listing the assets of the domain like –
- Subdomains from – Amass ,findomain, subfinder & resolvable subdomains using shuffledns
- Screenshots
- Port Scan
- JS files
- Httpx Status codes of subdomains
- Dirsearch file/dir paths by fuzzing
Installation
1 – Install Docker & docker-compose according to you OS from here – https://docs.docker.com/get-docker/
2 – git clone https://github.com/0xdekster/ReconNote.git
3 – Open docker-compose.yml & change the volumes directory path to the output folder
example –
volumes: – /root/reconnote/output/:/var/www/html
4 – Change the API_HOST parameter value to your server/host ip or domain name.
5 – Run docker-compose build OR docker-compose build –no-cache
6 – Run docker-compose up -d
7 – Reconnote framework will be up at – {your-server}:3000
Set Amass Config File to set API Keys
1- cd /ReconNote
2- docker exec -it reconnote_dekster_1 bash
3- cd /deksterrecon
4- nano amass-config.ini
5- Set your API keys and save, exit.
Usage
1 – Just enter domain/target name in Add Target & choose scan type
2 – Everything will be done by Reconnote and in few minutes you will get the Scan Results

Scan Result

Demo Video
Contributions
This is an open source project so contributins are welcome. You can request a PR for any changes that can enhance the ReconNote framework be it UI enhancement , tools adjustment ,features , etc..
Acknowledgements
ReconNote Security framework have been created by using the open source security tools made by amazing security community –
1- Eduard Tolosa
2- Tomnomnom
3- Michen riksen
4- Project Discovery
5- Corben Leo
Leave a Reply