BlueSpy is an Open Source Security Analysis Tool, works with GitHub, NPM and PyPi.
- Popularity Score
Takes into account the repo's forks, stars, watches and downloads (for NPM and PyPi) to create a single score. - Metadata Analysis Score
Metadata analysis module consists of 13 tests, 3 for Typosquatting and 10 for analyzing metadata. - Static Source Analysis
Uses Regex patterns to find common functionalities of malicious programs - cryptography, network calls, file system I/O, OS calls and cryptocurrency demands. - BlueSpy score
It's an aggregated score of the previous 3 scores that helps cluster packages into whether they are malicious or not. Generally, we found:- < 5: dangerous,
- 5-7: require manual code reviews before use,
- 7+: can be considered safe.
The backend server is a Docker container and requires GitHub Auth Token as an .env file:
GITHUB_AUTH_TOKEN=<TOKEN>
Then the image can be built and a container can be started listening on port 3000 using Docker:
docker build -t <IMAGE_NAME> .
docker run --rm -p 3000:3000 ---env-file=.env <IMAGE_NAME>
The frontend can be started using:
npm i && npm start