What problem does it solve?
Manual code quality checks are inconsistent, easy to skip, and AI coding agents often bypass linter rules by modifying configuration files instead of fixing actual code, leading to accumulating technical debt and unreliable codebases.
Core Features & Use Cases
- Write-time automatic enforcement: Runs formatters and linters on every file edit via Claude Code hooks, catching issues before they are committed, with support for Python, TypeScript, Shell, YAML, JSON, TOML, Markdown, and Dockerfile.
- Tiered AI-powered fixes: Routes code violations to the appropriate Claude model tier (Haiku for simple style fixes, Sonnet for complex refactoring, Opus for type system errors) to automatically resolve issues without manual developer intervention.
- Config tamper protection: Blocks edits to linter configuration files and detects unauthorized changes at session end to prevent agents from disabling rules to avoid fixing code.
Use case: For a development team using Claude Code, this skill resolves 40-50% of code quality issues silently, and automatically fixes most remaining violations, reducing time spent on manual linting and trivial code review tasks.
Quick Start
Use the plankton-code-quality skill to automatically enforce code formatting and linting rules on every file you edit while working in your project directory.
