Advanced Claude Code: Subagents, Hooks and Skills
This workshop is for teams that already use Claude Code and want to make it consistent and automatable. You will build subagents for focused roles, hooks that enforce your rules, and skills that package team knowledge. We work on your own repository so the results can be adopted straight away, with human review remaining the final gate.
Request without obligation- Duration
- 1 day
- Level
- Advanced
- Format
- Remote or on-site, English or German
- Group size
- Up to 12 participants per trainer
Who it is for
- Developers who use Claude Code daily
- Platform and developer-experience engineers
- Tech leads who want shared team conventions for agents
What your team will be able to do
- Design subagents with narrow roles and clear hand-offs
- Implement hooks that run checks or block risky actions automatically
- Package recurring procedures as reusable skills
- Structure project memory so agents follow team conventions
- Decide when to automate and when to keep a human in the loop
Agenda
- Review of the agent modelA quick recap of context, tools and permissions, and where single-agent sessions reach their limits.
- SubagentsSplitting work into specialised roles such as researcher, implementer and reviewer, and controlling their context and tool access.
- HooksRunning scripts before and after agent actions to format code, run tests or refuse unsafe commands.
- SkillsWriting reusable skills for team procedures such as release notes, migrations or test conventions.
- Project memory and conventionsOrganising instruction files so that every session starts with the right knowledge.
- Integration labCombine subagents, hooks and skills into one workflow for a real task and evaluate the result with the trainer.
Prerequisites
- Daily practical experience with Claude Code or a comparable agent
- Comfort with the shell and scripting
- A repository where you may add configuration files
Why apprat.io workshops
- Taught by practising AI Code Engineers
- Your repositories or a sandbox project
- AI speed, human responsibility: review, security and governance included
- ISO 9001 & ISO/IEC 27001 certified processes
- Follow-up office hours after 30 days
Request without obligation
Tell us about your team and goals. We reply within one business day with a tailored agenda and a quote – free and without obligation.
24/7 call support in English & German:
+49 761 519 751 50
Frequently asked questions
Does the workshop cover only Claude Code?
The patterns are discussed in general terms, but all exercises use Claude Code features. We map them to other tools where it helps.
Will we build something we can keep?
Yes. Subagent, hook and skill definitions created in your repository stay with you.
How do you handle security concerns with hooks?
We treat hooks as code that needs review, and we discuss least-privilege settings and sandboxing as part of the workshop.
Fields marked * are required.