This repository is a collection of reports of malicious packages identified in Open Source package repositories, consumable via the Open Source Vulnerability (OSV) format.
This project is closely related to the OpenSSF Package Analysis project.
Attacks against open source ecosystems are gaining popularity. Typosquatting, dependency confusion, account takeovers, etc are happening more frequently each year.
While some protection can be found through various security solutions, there is no comprehensive database of malicious packages published to open source package repositories.
The aim of this project and repository is to be a comprehensive, high quality, open source database of reports of malicious packages published on open source package repositories.
These public reports help protect the open source community, and provide a data source for the security community to improve their ability to find and detect new open source malware.
What is in scope?
- any package that belongs to an ecosystem supported by the OSV Schema
- malicious packages published under typosquatting or dependency confusion type attacks
- malicious packages published through account takeover
- prebuilt binaries for a package that are malicious
- security researcher activity
Out-of-scope:
- vulnerability reports
- compromised infrastructure
- non-malicious packages
- GitHub's Advisory Database (filtered by malware), for the NPM ecosystem.
- https://github.com/lxyeternal/pypi_malregistry (PyPI)
- https://dasfreak.github.io/Backstabbers-Knife-Collection/ (PyPI and npm), by Marc Ohm et al.
- https://github.com/datadog/malicious-software-packages-dataset (PyPI), by Datadog
See our contributing guide for complete details.
We accept new reports, and updates to existing reports.
We will also accept bulk imports via PR (please create an issue first).
If you regularly produce high-quality detections with few false-positives, and have them accumulating in a database, we can automatically consume them as OSV from a cloud storage environment (S3, GCS).
- Most communication occurs in the OpenSSF Package Analysis Slack channel
- Official communications occur on the https://lists.openssf.org/g/openssf-wg-securing-crit-prjs mailing list.
Manage your subscriptions to Open SSF mailing lists.
- Every other Thursday @ either an APAC or EMEA friendly time (See shared calendar).
- Meeting Minutes.
While we do our best to ensure false positives are not present, they may be present in our dataset from time-to-time.
If you see a non-malicious package is flagged as malicious create an issue. Please include the following:
- The affected ecosystem and package.
- Which versions are false positives, if specific versions are false positives.
- Any relevant links.
We will then either:
- Move the entire report into the
./osv/withdrawn/directory and add thewithdrawntime to the report - if the whole report is a false positive. - Move the affected versions into a
database_specificarray indicating that which versions were false positives - if some versions are malicious and some are false positives.
Note: support for handling false positives is TBC.
This work is associated with the Package Analysis project.
This project belongs to the Securing Critical Projects Working Group in the OpenSSF (Slack).
The working group's CHARTER.md outlines the scope and governance of our group activities.