Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
Pierre Clisson avatar

Wf

  • Updated March 12, 2026
  • mesca/claude-plugins

wf is a Claude Code skill in the Automation & Workflows category. AI-assisted workflows for Python development

Key points

  • wf
  • Automation & Workflows
  • AI-coding skill

Wf by the numbers

  • Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add mesca/claude-plugins
/plugin install wf@mesca

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Last updatedMarch 12, 2026
Repositorymesca/claude-plugins

What it does

AI-assisted workflows for Python development

README.md

Claude Plugins

Opinionated Python development standards for Claude Code. Scaffolds projects, enforces coding conventions, and audits compliance automatically.

Constitution

Core principles injected at every session start. Non-negotiable.

Development loop — always follow this sequence, never skip steps:

contract → models → tests → implementation → simplify → documentation → audit

Interface design — every interface (CLI, API, web) must be designed from the end-user perspective: obvious mental model, simple, intuitive, consistent, explicit.

Commands

User-invokable slash commands that run as autonomous subagents.

Command Description
/wf:init <name> [desc] Scaffold a new Python project with full structure and tooling
/wf:audit [category] Audit codebase compliance against development guidelines
/wf:simplifier [scope] Refine Python code for clarity and consistency

Usage examples:

/wf:init my-tool "A CLI tool for data processing"
/wf:simplifier
/wf:audit security
/wf:audit code testing

Background Skills

These skills are applied automatically by Claude when relevant:

Skill Purpose
conventions Python coding standards: structure, typing, logging, style
simplifier Code refinement for clarity and consistency
spec-driven Contracts-first development with OpenAPI/OpenRPC/docopt and models-first workflow
tdd Test-driven development: red-green-refactor with pytest
security Secrets management, input validation, injection prevention
performance Profiling, caching, memory optimization, async patterns
documentation Material for MkDocs setup and API doc generation
git-workflow Commit conventions, branch naming, pre-commit checks
project-name Detects and normalizes ${PROJECT_NAME} from pyproject.toml

Hooks

Hook Event Description
load-constitution.sh SessionStart Injects constitution into every session
notify.sh Notification Sends alerts to Telegram, Discord, and/or Slack

Project Architecture

Projects scaffolded by /wf:init follow this layered structure:

src/<package>/
    contracts/         # API schemas (OpenAPI 3.1, OpenRPC 1.3, docopt) — single source of truth
    core/              # Infrastructure only: logger, config, exceptions
    models/            # Shared Pydantic domain models
    services/          # Business logic (framework-agnostic)
    interfaces/        # Thin wrappers: cli/ (Typer), rest/ (FastAPI), rpc/ (JSON-RPC/WebSocket)

Tech Stack

Aspect Tool
Python version 3.13+
Package manager uv
Build system Hatchling + hatch-vcs
Linter/Formatter Ruff
Type checker Pyright (strict)
Test framework pytest
Property-based testing hypothesis
Documentation Material for MkDocs + mkdocstrings
CLI framework Typer
REST API FastAPI + uvicorn
JSON-RPC / WebSocket jsonrpc-websocket, websockets
Simple websites FastAPI + Jinja2 + HTMX
Logging loguru

Setup

Prerequisites

System Dependencies

macOS (via Homebrew):

brew install node uv

Linux (Debian/Ubuntu):

# Node.js (via NodeSource)
curl -fsSL https://deb.nodesource.com/setup_lts.x | sudo -E bash -
sudo apt-get install -y nodejs

# uv
curl -LsSf https://astral.sh/uv/install.sh | sh

Linux (Fedora/RHEL):

sudo dnf install nodejs
curl -LsSf https://astral.sh/uv/install.sh | sh

Verify installations:

node --version && npx --version && uv --version && uvx --version
Companion Plugins

Install from the official Anthropic Claude Plugins marketplace (run inside Claude Code):

/plugin install serena@claude-plugins-official
/plugin install context7@claude-plugins-official
/plugin install github@claude-plugins-official
/plugin install playwright@claude-plugins-official
/plugin install claude-md-management@claude-plugins-official
/plugin install ralph-loop@claude-plugins-official

These plugins configure their MCP servers automatically.

The github plugin requires a GITHUB_PERSONAL_ACCESS_TOKEN environment variable:

  1. Create a token at https://github.com/settings/tokens (Fine-grained or Classic)
  2. Add to your shell profile (~/.zshrc or ~/.bashrc):
    export GITHUB_PERSONAL_ACCESS_TOKEN="ghp_..."
    
  3. Reload your shell (source ~/.zshrc) before launching Claude Code.
Notifications (optional)

The notify.sh hook sends alerts when Claude Code needs attention (permission prompts, idle, etc.). Set any combination of these environment variables in your shell profile (~/.zshrc or ~/.bashrc):

export NOTIFY_TELEGRAM="bot_token|chat_id"
export NOTIFY_DISCORD="https://discord.com/api/webhooks/..."
export NOTIFY_SLACK="https://hooks.slack.com/services/..."

Telegram:

  1. Message @BotFather, send /newbot, follow the prompts to get a bot token
  2. Add the bot to your chat/group, then get the chat ID via https://api.telegram.org/bot<TOKEN>/getUpdates
  3. Format: bot_token|chat_id (e.g. 123456:ABC-DEF|-1001234567890)

Discord:

  1. In your server, go to Settings > Integrations > Webhooks > New Webhook
  2. Copy the webhook URL

Slack:

  1. Create an app at api.slack.com/apps, enable Incoming Webhooks
  2. Add a webhook to a channel and copy the URL
CC-SDD (Spec-Driven Development)

Adds /kiro:* commands for structured requirements → design → tasks → implementation workflow.

cd your-project
npx cc-sdd@latest --claude-agent --lang en

See https://github.com/gotalab/cc-sdd for all options (languages, agents, customization).

Install the Plugin

Inside Claude Code:

/plugin marketplace add mesca/claude-plugins
/plugin install wf@mesca

Restart Claude Code to load the plugin.

Project Setup

For new projects, use the init command:

/wf:init my-project "My project description"
cd my-project
uv sync --all-groups

For existing projects, ensure you have:

# Python 3.13+
uv python install 3.13

# Dev dependencies
uv add --dev pytest ruff pyright pip-audit hypothesis

TODO

  • UI skills
  • Docker skills
  • Git worktree rules

Related skills

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.