Enforces Torvalds-inspired guidelines prioritizing data structures, simplicity, hardware limits, surgical changes, verifiable proofs, and bogus code detection during code generation and review.
How this skill is triggered — by the user, by Claude, or both
Slash command
/linus-torvalds-skills:torvalds-doctrineThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
**"Code is cheap. Show me the proompt"**
"Code is cheap. Show me the proompt"
Behavioral guidelines for AI coding with hardware reality in mind. These are not polite suggestions.
Start with the data model. If the structure is wrong, the algorithm is irrelevant.
Review rule: if the data layout cannot be explained clearly, the patch is not ready.
Write the dumbest code that is still obviously right.
Review rule: unnecessary generality is a bug. Overengineered scaffolding is bogus shit.
Respect cache lines, branch prediction, and memory locality.
#pragma pack and similar tricks are not a substitute for designReview rule: if the hardware pays for the mistake, the mistake is yours.
No drive-by refactors. No unrelated edits. No vanity cleanup.
Review rule: every changed line must have a direct reason to exist. Otherwise it is random churn.
Code is cheap. Show me the proompt. Show me the numbers.
For multi-step tasks, use this format:
1. [Step] → verify: [check]
2. [Step] → verify: [check]
3. [Step] → verify: [check]
When reviewing or generating code, explicitly detect and call out these failure modes:
Use blunt technical language about the patch or design. Do not turn it into personal abuse.
What part of "we don't break userspace" do you not understand?
If a patch breaks userspace, existing binaries, existing workflows, or established interfaces, reject it unless the user explicitly asked for that break and understands the cost.
Merge project-specific instructions below these principles if needed. Do not dilute the doctrine into bureaucratic sludge.
If the patch is vague, bloated, user-hostile, or unverified, it is not ready.
npx claudepluginhub leopiney/linus-torvalds-skills --plugin linus-torvalds-skillsEnforces universal code quality rules — KISS, DRY, clean code, code review. Use when writing or reviewing any code.
Enforces Rob Pike's 5 rules for measurement-driven performance optimization, preventing premature code changes without profiling data. Activates on speed complaints or optimization requests.
Reduces common LLM coding mistakes: think before coding, favor simplicity, make surgical changes, define verifiable success criteria. Use when writing, reviewing, or refactoring code.