# AI Coding Tip 027 - Force Code Standards > TL;DR: Wire your standards into hooks, skills, and a judge, so the harness blocks violations before a human opens the diff. # Common Mistake ❌ You paste your coding standards into [AGENTS.md](https://maximilianocontieri.com/ai-coding-tip-011-initialize-agents-md) and trust the AI to remember them. Then [a human reviewer manually checks](https://maximilianocontieri.com/ai-coding-tip-006-review-every-line-before-commit) [naming](https://maximilianocontieri.com/code-smell-159-mixedcase), [indentation](https://maximilianocontieri.com/code-smell-164-mixed-indentations), and [spacing](https://maximilianocontieri.com/code-smell-211-tab-over-spaces) on every pull request. Nobody wires a hook, [a skill](https://maximilianocontieri.com/ai-coding-tip-004-use-modular-skills), or a judge that blocks the diff before a human ever opens it. The standards live in prose, and prose is optional. # Problems Addressed 😔 - A [CodeRabbit analysis of 470 pull requests](https://www.coderabbit.ai/blog/state-of-ai-vs-human-code-generation-report) found AI-generated code carries roughly 1.7x more defects than human-only code, and nearly 2x more naming and style consistency errors - [Human reviewers burn attention on indentation and casing instead of design](https://maximilianocontieri.com/ai-coding-tip-006-review-every-line-before-commit), feeding the same [broken windows](https://maximilianocontieri.com/code-smell-173-broken-windows) that erode a codebase over time - Enforcement depends on someone remembering to check, so it drifts the moment nobody's watching - The AI violates a rule it read in prose, because prose isn't a gate, only a suggestion - Two human reviewers catch two different subsets of the same [code without standards](https://maximilianocontieri.com/code-smell-48-code-without-standards) # How to Do It 🛠ī¸ 1. Turn each standard into a machine-checkable rule instead of a paragraph of prose. 2. Wire a pre-commit or pre-merge hook that runs the linter automatically on every change. 3. Add a validator skill that [reads the rule set fresh](https://maximilianocontieri.com/ai-coding-tip-005-keep-context-fresh) and re-checks the whole diff, not just what the AI remembers. 4. Route ambiguous rules, like naming intent, to a large language model (LLM) judge instead of a human reviewer. 5. Block the commit, the merge, or [the session end](https://maximilianocontieri.com/ai-coding-tip-024-force-a-criteria-check-before-the-task-ends) until every gate reports zero violations. 6. [Log every violation the judge catches as a new rule](https://maximilianocontieri.com/ai-coding-tip-016-feed-your-pr-lessons-into-the-ai-brain), so the harness never misses it again. # Benefits đŸŽ¯ 1. **Consistent gate:** Every diff hits the same rule set, so two reviewers never catch two different subsets of the same violation. 2. **Faster reviews:** [Humans spend their attention on design and correctness](https://maximilianocontieri.com/ai-coding-tip-006-review-every-line-before-commit), not indentation or casing. 3. **No memory decay:** A skill [reads the standards file fresh every session](https://maximilianocontieri.com/ai-coding-tip-005-keep-context-fresh) instead of trusting what the AI claims to remember. 4. **Judge for nuance:** An LLM judge catches the semantic violations a regular expression can't parse, like a misleading name. 5. **Compounding rules:** [Every violation the judge finds becomes a new check](https://maximilianocontieri.com/ai-coding-tip-025-pair-every-skill-with-a-pitfalls-file), so the harness gets stricter over time. 6. **Measurable drop:** You can track the naming and style defect rate over time and watch it fall toward zero. # Context 🧠 Standards enforcement isn't new. Stephen C. Johnson wrote the first lint in 1978 to catch mistakes in C code nobody wanted to check by eye. Checkstyle, PMD, and ESLint followed, each one refusing to trust a human to notice a [mixed indentation](https://maximilianocontieri.com/code-smell-164-mixed-indentations). [SonarQube](https://www.sonarsource.com/resources/library/linting-is-not-all-you-need/) added a server that gates a whole pipeline, not just a single file. [Mago](https://mago.carthage.software/) does the same for PHP now, running a linter, a formatter, and a static analyzer in one Rust binary fast enough to run on every keystroke. None of these tools ever asked a human to vote on a [tab versus a space](https://maximilianocontieri.com/code-smell-211-tab-over-spaces) or a [mixedCase versus snake_case](https://maximilianocontieri.com/code-smell-159-mixedcase) name. AI coding didn't remove that need. It multiplied it. A [CodeRabbit analysis of 470 pull requests](https://www.coderabbit.ai/blog/state-of-ai-vs-human-code-generation-report) found AI-generated code carries about 1.7x more defects overall than human-only code. Naming and style consistency errors came in at nearly 2x, the exact class of mistake a linter was built to catch decades ago. A model can now write in [any human language](https://maximilianocontieri.com/code-smell-128-non-english-coding), [misspell an identifier](https://maximilianocontieri.com/code-smell-98-speling-mistakes), or [reorder parameters inconsistently](https://maximilianocontieri.com/code-smell-87-inconsistent-parameters-sorting) across two files it never compared side by side. A linter still catches the syntactic version of these mistakes. The semantic version, like a name that lies about its role, needs a judge that [reads intent, not just tokens](https://maximilianocontieri.com/ai-coding-tip-019-tell-the-ai-why-not-just-what). That's where an [LLM-as-judge](https://arxiv.org/pdf/2508.02994) step fits: a second AI pass, wired into the same [harness](https://maximilianocontieri.com/ai-coding-tip-022-give-ai-a-harness-to-work-with) that runs the linter, checking the rules no regular expression can express. [Skills](https://maximilianocontieri.com/ai-coding-tip-004-use-modular-skills) are the natural home for that judge. A skill reads the standards file fresh every time, instead of trusting a memory that decays across [sessions](https://maximilianocontieri.com/ai-coding-tip-005-keep-context-fresh). Pair the judge with a [criteria check before the task ends](https://maximilianocontieri.com/ai-coding-tip-024-force-a-criteria-check-before-the-task-ends), so the AI can't say a task is done until the standards gate reports zero violations. A gate that blocks the diff is just another way to [force the AI to obey you](https://maximilianocontieri.com/ai-coding-tip-015-force-the-ai-to-obey-you), not by asking nicely, but by refusing to let a violation through. This doesn't replace [reviewing every line before commit](https://maximilianocontieri.com/ai-coding-tip-006-review-every-line-before-commit). It removes the mechanical part of that review, so [a human is free to judge design instead of counting spaces](https://maximilianocontieri.com/ai-coding-tip-006-review-every-line-before-commit). When the judge catches a new violation, log it the same way you'd log a [pitfall](https://maximilianocontieri.com/ai-coding-tip-025-pair-every-skill-with-a-pitfalls-file), so the harness never makes that mistake twice. ## Prompt Reference 📝 ## Bad Prompt đŸšĢ ```markdown Please follow our coding standards for this feature. Use consistent naming and formatting like the rest of the codebase. I'll check it during code review before we merge. ``` ## Good prompt 👉 ```markdown Before you say this task is done, run this standards checklist: - [ ] Run the linter on every changed file, fix every violation. - [ ] Run the formatter, don't hand-format a single line. - [ ] Invoke the code-standards-validator skill on the full diff. - [ ] Check every identifier: no abbreviations, no misleading names. - [ ] Check indentation matches the project config, no mixed tabs. - [ ] Check casing: one convention, no mixedCase next to snake_case. - [ ] Check parameter order against other functions in the file. - [ ] Check spelling in every identifier, comment, and string. - [ ] Judge comment quality: flag dead comments and restated code. - [ ] Judge naming intent: does each name say what it does? - [ ] Confirm no file mixes languages in identifiers or comments. - [ ] Log any new violation type as a rule for the next run. - [ ] Don't report the task done until every box above is checked. Show me the completed checklist, not just the final code. ``` # Considerations ⚠ī¸ A linter still beats an LLM judge on speed and cost for anything syntactic. Reserve the judge for rules that need intent, not tokens. A gate that blocks too aggressively gets bypassed with `--no-verify`, which defeats the whole point. Review the judge's false positives the same way you'd review a flaky linter rule. # Type 📝 [X] Semi-Automatic # Limitations ⚠ī¸ An LLM judge costs tokens and time on every gate, so it doesn't replace a linter. It complements one. A judge can disagree with itself across two runs on borderline style calls, so keep the deterministic rules in the linter and reserve the judge for what's genuinely ambiguous. # Tags 🏷ī¸ - Knowledge Management # Level 🔋 [X] Intermediate # Related Tips 🔗 %[https://maximilianocontieri.com/ai-coding-tip-004-use-modular-skills] %[https://maximilianocontieri.com/ai-coding-tip-006-review-every-line-before-commit] %[https://maximilianocontieri.com/ai-coding-tip-011-initialize-agents-md] %[https://maximilianocontieri.com/ai-coding-tip-015-force-the-ai-to-obey-you] %[https://maximilianocontieri.com/ai-coding-tip-016-feed-your-pr-lessons-into-the-ai-brain] %[https://maximilianocontieri.com/ai-coding-tip-019-tell-the-ai-why-not-just-what] %[https://maximilianocontieri.com/ai-coding-tip-022-give-ai-a-harness-to-work-with] %[https://maximilianocontieri.com/ai-coding-tip-024-force-a-criteria-check-before-the-task-ends] %[https://maximilianocontieri.com/ai-coding-tip-025-pair-every-skill-with-a-pitfalls-file] # Conclusion 🏁 A linter never asked permission to reject bad code. Neither should your harness. Wire the standards into hooks, skills, and a judge, so the diff gets rejected before a human ever has to say so. # More Information ℹī¸ [Lint (software) - Wikipedia](https://en.wikipedia.org/wiki/Lint_(software)) [State of AI vs Human Code Generation Report - CodeRabbit](https://www.coderabbit.ai/blog/state-of-ai-vs-human-code-generation-report) [When AIs Judge AIs: The Rise of Agent-as-a-Judge Evaluation for LLMs](https://arxiv.org/pdf/2508.02994) [Awesome Static Analysis - curated list of linters and code quality tools](https://github.com/lukehutch/awesome-static-analysis) # Also Known As 🎭 - Harness-Enforced-Standards - Machine-Judged-Style - Standards-as-Code - Linter-Native-Review # Tools 🧰 [Mago](https://mago.carthage.software/) [ESLint](https://eslint.org/) [SonarQube](https://www.sonarsource.com/products/sonarqube/) [PHP_CodeSniffer](https://github.com/PHPCSStandards/PHP_CodeSniffer) [Ruff](https://docs.astral.sh/ruff/) [RuboCop](https://rubocop.org/) [Checkstyle](https://checkstyle.sourceforge.io/) [golangci-lint](https://golangci-lint.run/) [Clippy](https://doc.rust-lang.org/clippy/) # Disclaimer đŸ“ĸ The views expressed here are my own. I am a human who writes as best as possible for other humans. I use AI proofreading tools to improve some texts. I welcome constructive criticism and dialogue. I shape these insights through 30 years in the software industry, 25 years of teaching, and writing over 500 articles and a book. * * * This article is part of the *AI Coding Tip* series. %[https://maximilianocontieri.com/ai-coding-tips]