🤖

Formation · AI · Developer agent · Spec-Driven

Claude Code

From onboarding Claude Code to deploying a full agentic Spec-Driven Development workflow: install, prompts, skills, plugins, agents — then move to a reproducible cycle spec → plan → tasks → implement → review → merge with Spec Kitty.

Duration

3 days

Level

All levels

Format

Inter / Intra

Inter-company price

From €1,950

Students

10 max

Book a session Request a private session

Learning objectives

What you'll be able to do after the training.

⚡

Master the tool

Install, configure and use Claude Code daily in real development contexts.

🧩

Extend & customize

Design skills, plugins and agents tailored to your stack and team practices.

🛡

Adopt as a team

Define a secure usage framework, shared conventions and measure real benefits.

📐

Industrialize Spec-Driven

Set up a reproducible spec → plan → tasks → implement → review → merge cycle with Spec Kitty and Claude Code.

Target audience

Who it's for, what's required.

👥 Audience

Developers wanting to accelerate coding cycles
Lead Tech and Tech Leads driving team adoption
DevOps / SRE engineers for scripting and automation
Architects prototyping components or integrations

📋 Prerequisites

Regular software development practice
Familiarity with a terminal and the command line
Git versioning knowledge
Laptop with internet access for the labs

Detailed program

Syllabus, day by day.

Discovery & mastery

Day 1 — Get hands-on with Claude Code and its usage modes

Claude Code introduction and architecture
Installation, configuration, authentication
Interactive vs headless modes
Designing effective prompts
Context management: CLAUDE.md, memory, files
Slash commands, shortcuts, available models
Best practices for reviewing and validating outputs

LabRefactor an existing module and write its automated tests with Claude Code, starting from a real repo.

Industrialization & team

Day 2 — Extend, secure, deploy as a team

Skills: design, tests, distribution
Plugins: structure, marketplace, packaging
MCP (Model Context Protocol) configuration
Custom agents and automation hooks
Security, scopes, sensitive data, audit
Team conventions, ownership, reviews
Impact measurement and adoption indicators

LabDesign a custom skill or plugin reproducible for your team: business need, structure, test and publishing.

Spec-Driven Development with Spec Kitty

Day 3 — Ship features end-to-end

Why Spec-Driven Development: from prompt-and-pray to a traced workflow
Spec Kitty CLI: install, governance, missions
Full cycle: specify → plan → tasks → implement → review → merge
Drive Claude Code in a loop (single task or full feature)
Git worktrees, kanban dashboard, GitHub integration

LabShip a real feature end-to-end with Spec Kitty: init the project, write the spec, generate plan + tasks, run the Claude Code implementation loop, validate via review and merge.