I found 10k GitHub repositories distributing Trojan malware
Thousands of GitHub repositories are being used to spread Trojan malware via booby‑trapped binaries and archives, often masquerading as useful tools or coding test projects and likely aiming to steal cryptocurrency and account credentials. Commenters describe weak or inconsistent responses from GitHub and antivirus services, note how search engines and fake forks amplify the problem, and debate how much trust to place in open source when most users never audit code or verify binaries. Many call for stronger automated scanning, better ecosystem hygiene, and stricter personal security practices such as sandboxing, password‑manager hygiene, and hardware-based authentication.
Nature of the Malware Campaign
- Repositories host seemingly useful projects but distribute malware via linked ZIP archives or replaced binaries in “releases.”
- Samples show network activity to IP info services, Polygon RPC nodes, and a C2 server, suggesting crypto theft.
- One analyzed sample matches the “disco” trojan family; most payloads are Windows-focused.
- Attackers frequently delete and re-push commits to stay at the top of “recently updated” and search results.
GitHub’s Role and Response
- Mixed experiences reporting: some repos removed within 24 hours; others linger for months or years (including obvious piracy).
- Commenters argue GitHub is not a curated repository, more like a file host or forum with links.
- Several feel GitHub/Microsoft underuse their resources (including AI) for proactive malware detection, likely due to business tradeoffs and false-positive risks.
Limits of “Open Source Is Safe”
- Strong debate over the common belief that open source is “safer” because it’s auditable.
- Many note that:
- Few people actually read code or verify that binaries match sources.
- Stars, popularity, and “many eyes” create a false sense of security and are easily gamed.
- Others push back that open source was never guaranteed clean, only more auditable; xz backdoor cited as a counterexample to naïve trust.
Detection, Scanning, and Tooling
- VirusTotal sometimes flags the malware only when the ZIP is uploaded directly, not when given just a link.
- Calls for GitHub-wide release/binary scanning and basic heuristics (new accounts, synchronized stargazers, cloned malware trees).
- Existing tools mentioned: Socket, Aikido, Step-Security, Wiz, Semgrep-style scanning; some want better pre-download GitHub scanners.
Search Engines and Phishing
- Search results (especially non-Google) often surface phishing or malicious GitHub forks above legitimate repos.
- Some users report phishing pages for banks and GitHub projects ranking highly; others say this happens on multiple major search engines.
Account Security and Password Practices
- A high-profile case of malware from a GitHub plugin compromising a password manager drives debate over:
- Storing TOTP/MFA in the same vault as passwords.
- Using hardware keys (e.g., YubiKey), separate devices, or multiple vaults.
- Several argue password managers assume an uncompromised device; once malware runs locally, many defenses fail.
Social Engineering and Fake Technical Tests
- Descriptions of recruitment scams: “dream job” offers, then a “technical test” repo that actually deploys malware.
- Some developers now run all such tests only in disposable VMs and monitor outbound connections.
Broader Concerns About Ecosystem Quality
- Perception that GitHub is drifting toward “SourceForge 2.0”: spam, malware, SEO-gamed stars, and weak enforcement.
- Some see this as a long-standing open ecosystem reality; others blame large corporate ownership and growth incentives.