Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
The Torvalds Doctrine is a set of aggressive coding guidelines that turns an AI assistant into a disciplined reviewer. It enforces data structure supremacy, rejects overengineering, demands proof over hand-waving, and calls out bogus shit on sight. Developers who want their AI-generated code to stay grounded in hardware reality and avoid enterprise sludge will find a blunt counterpart in the chat.
WHO IT'S FOR
Software Engineers Using AI Coding Assistants
enforce code quality in AI patches
Technical Leads and Code Reviewers
establish review criteria for AI-generated contributions
Open-Source Maintainers
guard against low-quality AI patches in projects
Platform and DevEx Teams
embed coding standards into AI tooling configurations
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
A patch or diff path replacing [PATCH_OR_DIFF_PATH]
Any relevant compatibility constraints or expected behavior
Use the torvalds-doctrine skill to review the patch at [PATCH_OR_DIFF_PATH]. Identify up to three concrete problems involving the data model, unnecessary complexity, unrelated changes, compatibility regressions, or unsupported correctness/performance claims. For each problem, cite the relevant changed lines, explain the technical impact, recommend the smallest corrective change, and name a test, benchmark, or reproducible check that would verify it. Review the patch only; do not modify files.
Claude
Cursor