New to Claude Skills? Learn how to install them →

Paffaan-m on GitHub

Plankton Code Quality

Free

Automate code quality enforcement during development.

Get this skill

Free · Opens the source repo

What Plankton Code Quality does

Plankton is a code quality enforcement skill designed for use with Claude Code, enabling automatic formatting and linting of code at every file edit. This skill integrates a multi-stage process that runs formatting tools and linters through a PostToolUse hook, ensuring that code adheres to specified quality standards without requiring manual intervention. The system also utilizes Claude's capabilities to fix unaddressed issues, streamlining the development workflow.

The operation of Plankton is structured into three distinct phases. First, it automatically formats code silently, addressing 40-50% of issues without user notification. In the second phase, it collects remaining violations and returns them in a structured JSON format. Finally, it delegates complex issues to Claude for resolution, routing them based on their severity through a tiered model that optimizes response time and accuracy. This ensures that developers receive immediate feedback and remediation for code quality issues.

Plankton is particularly useful for teams that require consistent code quality across multiple languages including Python, TypeScript, Shell, YAML, JSON, TOML, Markdown, and Dockerfile. It is designed to prevent modifications to linter configurations that could compromise code quality, implementing protective measures against configuration tampering. This makes it ideal for environments where maintaining strict quality standards is critical.

For integration, Plankton requires minimal setup, automatically activating hooks when Claude Code is initiated. Developers can quickly configure it for their projects by copying necessary hook files and linter configurations. The skill also includes provisions for package management enforcement, ensuring that legacy package managers do not interfere with the development process, thus maintaining a clean and efficient coding environment.

When to use it

Use Plankton when you want to ensure code quality is maintained automatically during every file edit, not just at commit time.

When not to use it

Plankton may not be suitable for projects with minimal code quality requirements or where manual code review processes are preferred.

What you can build with it

Automated Code Formatting

Developers can rely on Plankton to automatically format their code with each edit, ensuring consistent style without manual effort.

Real-time Code Quality Checks

Plankton provides immediate feedback on code quality, allowing developers to address issues as they arise rather than during code reviews.

Multi-language Support

Teams working in diverse programming languages can use Plankton to enforce uniform code quality standards across all their projects.

How to install Plankton Code Quality

View source

1. Install with the skills CLI

npx skills add affaan-m/ecc/plankton-code-quality --agent claude-code

2. Or install it manually

Download the skill folder and drop it into ~/.claude/skills/ for all projects, or .claude/skills/ to scope it to one repo. Restart Claude Code so it picks up the new skill.

Anthropic's agentic coding CLI, and the reference implementation of Agent Skills. Drop a skill folder into ~/.claude/skills and Claude Code loads it automatically whenever a task matches the skill's description. Claude Code docs

Inside SKILL.md

Written by affaan-m

Plankton 代码质量技能

Plankton(作者:@alxfazio)的集成参考,这是一个用于 Claude Code 的编写时代码质量强制执行系统。Plankton 通过 PostToolUse 钩子在每次文件编辑时运行格式化程序和 linter,然后生成 Claude 子进程来修复代理未捕获的违规。

何时使用

  • 你希望每次文件编辑时都自动格式化和检查(不仅仅是提交时)
  • 你需要防御代理修改 linter 配置以通过检查,而不是修复代码
  • 你想要针对修复的分层模型路由(简单样式用 Haiku,逻辑用 Sonnet,类型用 Opus)
  • 你使用多种语言(Python、TypeScript、Shell、YAML、JSON、TOML、Markdown、Dockerfile)

工作原理

三阶段架构

每次 Claude Code 编辑或写入文件时,Plankton 的 multi_linter.sh PostToolUse 钩子都会运行:

阶段 1:自动格式化(静默)
├─ 运行格式化工具(ruff format、biome、shfmt、taplo、markdownlint)
├─ 静默修复 40-50% 的问题
└─ 无输出至主代理

阶段 2:收集违规项(JSON)
├─ 运行 linter 并收集无法修复的违规项
├─ 返回结构化 JSON:{line, column, code, message, linter}
└─ 仍无输出至主代理

阶段 3:委托 + 验证
├─ 生成带有违规项 JSON 的 claude -p 子进程
├─ 根据违规项复杂度路由至模型层级:
│   ├─ Haiku:格式化、导入、样式(E/W/F 代码)—— 120 秒超时
│   ├─ Sonnet:复杂度、重构(C901、PLR 代码)—— 300 秒超时
│   └─ Opus:类型系统、深度推理(unresolved-attribute)—— 600 秒超时
├─ 重新运行阶段 1+2 以验证修复
└─ 若清理完毕则退出码 0,若违规项仍存在则退出码 2(报告至主代理)

主代理看到的内容

场景代理看到钩子退出码
无违规0
全部由子进程修复0
子进程后仍存在违规[hook] N violation(s) remain2
建议性警告(重复项、旧工具)[hook:advisory] ...0

主代理只看到子进程无法修复的问题。大多数质量问题都是透明解决的。

配置保护(防御规则博弈)

LLM 会修改 .ruff.tomlbiome.json 来禁用规则,而不是修复代码。Plankton 通过三层防御阻止这种行为:

  1. PreToolUse 钩子protect_linter_configs.sh 在编辑发生前阻止对所有 linter 配置的修改
  2. Stop 钩子stop_config_guardian.sh 在会话结束时通过 git diff 检测配置更改
  3. 受保护文件列表.ruff.toml, biome.json, .shellcheckrc, .yamllint, .hadolint.yaml

包管理器强制执行

Bash 上的 PreToolUse 钩子会阻止遗留包管理器:

  • pip, pip3, poetry, pipenv → 被阻止(使用 uv
  • npm, yarn, pnpm → 被阻止(使用 bun
  • 允许的例外:npm audit, npm view, npm publish

设置

快速开始

# Clone Plankton into your project (or a shared location)
# Note: Plankton is by @alxfazio
git clone https://github.com/alexfazio/plankton.git
cd plankton

# Install core dependencies
brew install jaq ruff uv

# Install Python linters
uv sync --all-extras

# Start Claude Code — hooks activate automatically
claude

无需安装命令,无需插件配置。当你运行 Claude Code 时,.claude/settings.json 中的钩子会在 Plankton 目录中被自动拾取。

按项目集成

要在你自己的项目中使用 Plankton 钩子:

  1. .claude/hooks/ 目录复制到你的项目
  2. 复制 .claude/settings.json 钩子配置
  3. 复制 linter 配置文件(.ruff.toml, biome.json 等)
  4. 为你使用的语言安装 linter

语言特定依赖

语言必需可选
Pythonruff, uvty(类型), vulture(死代码), bandit(安全)
TypeScript/JSbiomeoxlint, semgrep, knip(死导出)
Shellshellcheck, shfmt
YAMLyamllint
Markdownmarkdownlint-cli2
Dockerfilehadolint (>= 2.12.0)
TOMLtaplo
JSONjaq

与 ECC 配对使用

互补而非重叠

关注点ECCPlankton
代码质量强制执行PostToolUse 钩子 (Prettier, tsc)PostToolUse 钩子 (20+ linter + 子进程修复)
安全扫描AgentShield, security-reviewer 代理Bandit (Python), Semgrep (TypeScript)
配置保护PreToolUse 阻止 + Stop 钩子检测
包管理器检测 + 设置强制执行(阻止遗留包管理器)
CI 集成用于 git 的 pre-commit 钩子
模型路由手动 (/model opus)自动(违规复杂度 → 层级)

推荐组合

  1. 将 ECC 安装为你的插件(代理、技能、命令、规则)
  2. 添加 Plankton 钩子以实现编写时质量强制执行
  3. 使用 AgentShield 进行安全审计
  4. 在 PR 之前使用 ECC 的 verification-loop 作为最后一道关卡

避免钩子冲突

如果同时运行 ECC 和 Plankton 钩子:

  • ECC 的 Prettier 钩子和 Plankton 的 biome 格式化程序可能在 JS/TS 文件上冲突
  • 解决方案:使用 Plankton 时禁用 ECC 的 Prettier PostToolUse 钩子(Plankton 的 biome 更全面)
  • 两者可以在不同的文件类型上共存(ECC 处理 Plankton 未覆盖的内容)

配置参考

Plankton 的 .claude/hooks/config.json 控制所有行为:

{
  "languages": {
    "python": true,
    "shell": true,
    "yaml": true,
    "json": true,
    "toml": true,
    "dockerfile": true,
    "markdown": true,
    "typescript": {
      "enabled": true,
      "js_runtime": "auto",
      "biome_nursery": "warn",
      "semgrep": true
    }
  },
  "phases": {
    "auto_format": true,
    "subprocess_delegation": true
  },
  "subprocess": {
    "tiers": {
      "haiku":  { "timeout": 120, "max_turns": 10 },
      "sonnet": { "timeout": 300, "max_turns": 10 },
      "opus":   { "timeout": 600, "max_turns": 15 }
    },
    "volume_threshold": 5
  }
}

关键设置:

  • 禁用你不使用的语言以加速钩子
  • volume_threshold — 违规数量超过此值自动升级到更高的模型层级
  • subprocess_delegation: false — 完全跳过第 3 阶段(仅报告违规)

环境变量覆盖

变量目的
HOOK_SKIP_SUBPROCESS=1跳过第 3 阶段,直接报告违规
HOOK_SUBPROCESS_TIMEOUT=N覆盖层级超时时间
HOOK_DEBUG_MODEL=1记录模型选择决策
HOOK_SKIP_PM=1绕过包管理器强制执行

参考

  • Plankton(作者:@alxfazio)
  • Plankton REFERENCE.md — 完整的架构文档(作者:@alxfazio)
  • Plankton SETUP.md — 详细的安装指南(作者:@alxfazio)

ECC v1.8 新增内容

可复制的钩子配置文件

设置严格的质量行为:

export ECC_HOOK_PROFILE=strict
export ECC_QUALITY_GATE_FIX=true
export ECC_QUALITY_GATE_STRICT=true

语言关卡表

  • TypeScript/JavaScript:首选 Biome,Prettier 作为后备
  • Python:Ruff 格式/检查
  • Go:gofmt

配置篡改防护

在质量强制执行期间,标记同一迭代中对配置文件的更改:

  • biome.json, .eslintrc*, prettier.config*, tsconfig.json, pyproject.toml

如果配置被更改以抑制违规,则要求在合并前进行明确审查。

CI 集成模式

在 CI 中使用与本地钩子相同的命令:

  1. 运行格式化程序检查
  2. 运行 lint/类型检查
  3. 严格模式下快速失败
  4. 发布修复摘要

健康指标

跟踪:

  • 被关卡标记的编辑
  • 平均修复时间
  • 按类别重复违规
  • 因关卡失败导致的合并阻塞

Frequently asked questions about Plankton Code Quality

Similar skills