Scanner methodology

What is slopsquatting?

Slopsquatting exploits package names invented by AI coding assistants. An attacker can register a suggested name on a package registry and publish malicious code. A developer who trusts the suggestion may then install the attacker’s package.

Slopsquatting vs typosquatting

Typosquatting uses names similar to existing packages to exploit spelling mistakes. Slopsquatting exploits names that an AI model suggests, whether or not they resemble a real package. A suspicious name alone does not establish that an AI invented it or that its contents are malicious.

What does this scanner check?

Paste a public GitHub repository URL into the scanner. It reads supported dependency declarations in the root package.json and pyproject.toml on the default branch. It uses npm and PyPI registry metadata, cached results, and a built-in list of well-known packages to identify names that need investigation.

Signals include missing packages, security holds, deprecation, package age, release history, download counts, and source repository availability. Results include the manifest location and an explanation of each finding.

What are the limitations?

This is a dependency-name and registry-metadata check. It does not execute or audit package payloads, recursively scan every workspace or transitive dependency, or automatically block an installation. Private packages, unpublished workspace dependencies, registry outages, and new legitimate packages can require manual review. A clean result is not a guarantee that a repository is safe.

What should I do with a flagged package?

Check the exact name against the project’s official installation instructions. Review the registry entry, source repository, maintainers, and release history before installing it. Use this scanner alongside dependency auditing, lockfile review, and your existing security checks.

Browse flagged npm and PyPI packages to review findings collected by the scanner.