How does Plankton handle unfixable linting errors?

Plankton identifies unfixable violations and spawns a background Claude subprocess (using Haiku, Sonnet, or Opus depending on complexity) to attempt a fix before the main agent session continues.

Which programming languages are supported?

Plankton supports a wide array of languages including Python (Ruff), TypeScript/JS (Biome), Shell (ShellCheck), YAML, JSON, TOML, Markdown, and Dockerfiles.

What is the benefit of write-time enforcement over CI/CD?

Write-time enforcement catches and fixes errors the moment they are created, ensuring the main agent never proceeds with broken code and reducing the feedback loop compared to waiting for a CI build.

Can an AI agent disable my linter rules to bypass checks?

No, Plankton includes dedicated PreToolUse hooks and a config guardian that specifically block agents from modifying files like .ruff.toml or biome.json.

Plankton Code Quality & Auto-Fix

Name: Plankton Code Quality & Auto-Fix
Author: affaan-m

byaffaan-m

•

130,863

•

开发者工具

Enforces strict code quality through automated formatting, multi-language linting, and intelligent AI-powered remediation on every file edit.

Plankton is a comprehensive write-time quality enforcement system designed to maintain production-grade standards within Claude Code environments. It operates via a three-phase architecture that automatically formats code, identifies linting violations, and intelligently routes unfixable errors to specialized Claude subprocesses (Haiku, Sonnet, or Opus) for silent remediation. By implementing strict configuration protection, it prevents AI agents from 'gaming' the system by disabling rules, while also enforcing modern package managers like uv and bun to ensure codebase consistency across Python, TypeScript, Shell, and more.

主要功能

01Automated multi-phase formatting and linting triggered on every file modification

02Legacy package manager blocking to enforce modern tools like uv and bun

03130,863 GitHub stars

04Tiered AI model routing for automated error remediation based on violation complexity

05Config Tamper Guard to prevent agents from modifying or disabling linter rules

06Native support for Python, TypeScript, Shell, YAML, Dockerfile, and Markdown

使用场景

01Automating complex type-checking and refactoring tasks during the development loop

02Maintaining architectural and style consistency in polyglot repositories without manual reviews

03Preventing AI agents from introducing technical debt or bypassing security linters

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add affaan-m/everything-claude-code plankton-code-quality

For use in Claude.ai and ChatGPT

主要功能

01Automated multi-phase formatting and linting triggered on every file modification

02Legacy package manager blocking to enforce modern tools like uv and bun

03130,863 GitHub stars

04Tiered AI model routing for automated error remediation based on violation complexity

05Config Tamper Guard to prevent agents from modifying or disabling linter rules

06Native support for Python, TypeScript, Shell, YAML, Dockerfile, and Markdown

使用场景

01Automating complex type-checking and refactoring tasks during the development loop

02Maintaining architectural and style consistency in polyglot repositories without manual reviews

03Preventing AI agents from introducing technical debt or bypassing security linters