Torvalds Doctrine

Linus-style code-review doctrine: data-structure-first, kill over-engineering, demand proof

Tested · Works

Test report

Verdict
Tested · Works
Score
8.0/10
Tested
Jul 17, 2026
Environment
Claude Code 2.x (agent harness)
Upstream re-checked
Aug 10, 2026 · 5386e1d

Reviewed a deliberately over-engineered 22-line snippet both ways: the doctrine's pass led with a concrete data-structure fix (replace the size elif-chain with a threshold table), demanded a benchmark for the unproven '2x faster' comment, and named the unreachable else + redundant n>=1000 branch as special-case insanity, where the no-skill baseline flagged the same issues but only as a vague 'could simplify'. A modest but real gain in actionability; it is a guideline prompt, not an enforcer.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

  • Installs cleanly 5/5
  • Triggers reliably 5/5
  • Output vs. baseline 6/10
  • Docs & honesty 4/5

What Torvalds Doctrine does

A behavioral code-review skill that makes the model lead with the data structure, reject speculative abstractions and drive-by refactors, and demand benchmarks for performance claims. Includes a named 'bogus-shit detector' checklist of failure modes. Triggers on requests for a blunt, opinionated code or patch review.

How to install Torvalds Doctrine

git clone https://github.com/leopiney/linus-torvalds-skills
mkdir -p ~/.claude/skills
cd linus-torvalds-skills && cp -r skills/torvalds-doctrine ~/.claude/skills/torvalds-doctrine

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Torvalds Doctrine

  • /torvalds-doctrine Linus-style code-review doctrine: data-structure-first, kill over-engineering, demand proof

It also activates on plain-language prompts like these:

  • Give me a blunt, no-nonsense review of this patch
  • Rip apart this over-engineered snippet and tell me what's wrong
  • Review this pull request the way Linus Torvalds would

Frequently asked questions

Is the Torvalds Doctrine skill free?
Yes. The skill itself is free from leopiney/linus-torvalds-skills. SkillProof publishes the install command and an independent test verdict at no cost.
Does Torvalds Doctrine work with Claude Code?
We tested it with Claude Code 2.x (agent harness) on Jul 17, 2026. Verdict: Tested · Works. Reviewed a deliberately over-engineered 22-line snippet both ways: the doctrine's pass led with a concrete data-structure fix (replace the size elif-chain with a threshold table), demanded a benchmark for the unproven '2x faster' comment, and named the unreachable else + redundant n>=1000 branch as special-case insanity, where the no-skill baseline flagged the same issues but only as a vague 'could simplify'. A modest but real gain in actionability; it is a guideline prompt, not an enforcer.
What is the Torvalds Doctrine SkillProof Score?
8.0/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 6/10, docs & honesty 4/5.
How do I install Torvalds Doctrine?
Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Torvalds Doctrine with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.