TrustOSS

TrustOSS helps developers evaluate GitHub dependencies before installing them. Users can search for libraries by keyword or paste an owner/repo or GitHub URL to get a health report. The overall health score combines five weighted dimensions (activity, maintenance, community, documentation, and maturity) computed from public GitHub data. A fake-star audit plots star growth and profiles recent stargazers, and a deep scan reports OK, WARN, or RISK findings covering vulnerabilities from OSV, dependents from deps.dev, bus factor, CI pass rate, PR merge rate, and docs-only detection. Reports are shareable, comparisons can be made across 2 to 4 libraries, and an embeddable SVG README badge displays a repository's live score. A CLI (trustoss-cli) is also offered, which includes an agent skill for AI coding agents. Searching, analyzing, comparing, deep scans, star audits, and badges are stated to be free and require no account; GitHub sign-in is only needed for the maintainer dashboard.

Key features

• Health score from 0 to 100 with letter grade across five weighted dimensions: activity, maintenance, community, documentation, and maturity
• Fake-star audit that plots star growth curve and profiles recent stargazers by account age, followers, and repositories
• Deep scan with known vulnerability checks from OSV, dependent packages from deps.dev, bus factor, CI pass rate, PR merge rate, and docs-only detection
• Side-by-side comparison of 2 to 4 libraries with radar chart and score table, highlighting the strongest pick
• Embeddable SVG README badge showing live TrustOSS score
• Shareable report URLs at /repo/owner/name
• Command-line interface (trustoss-cli) with analyze, deep-scan, and star-audit commands
• Agent skill installed by the CLI so AI coding agents can vet dependencies

Best suited for

• Developers evaluating open-source dependencies before adding them to a project
• Teams assessing bus factor, maintenance, and vulnerability exposure of libraries
• Maintainers who want to display a TrustOSS score badge in their README
• Users comparing candidate libraries for a feature need

Pricing

Free; searching, analyzing, comparing, deep scans, star audits, and badges require no account (page states it is free)