Agentic Workflows
The earlier posts in this series walked through the basics of agentic engineering. This one is more personal: a tour of the workflow I actually run day to day. The core idea is simple — skills package up a full workflow so I can reuse it, hooks automate the small things around it, and a handful of daily tools keep input, monitoring, and output fast. Combined, they make my work both more stable and more efficient.
Skills: Reusable Workflow Packages
What a skill is, why “reusable” matters, and how it compares to a one-off prompt.
How I Structure a Skill
The anatomy of one of my skills — name, trigger, body. One concrete example walked through end to end.
Skills I Reach for Daily
A short list (3–5) of the skills I use most, with one sentence each on what they do for me.
Hooks: Small Automations That Add Up
What a hook is in Claude Code, and why I treat it as the “glue” around skills rather than a workflow on its own.
Hooks I Have Set Up
A short list of the hooks I run (e.g., format-on-save, lint-on-stop, custom guards) with one sentence each on what they prevent or accelerate.
Skills × Hooks: Stability and Efficiency
The combination effect. Skills give me a repeatable starting point; hooks keep the guardrails on. One short story of a task where the two together saved me from a mistake or a long redo.
Daily Tools
Software outside Claude Code that supports the loop. Three sub-buckets:
Faster Input
Voice dictation, snippet expanders, custom keyboards — whatever shortens the path from idea to text.
Progress Monitoring
How I keep an eye on long-running agents, tasks, or builds without babysitting them.
Output Generation
Tools that take the agent’s output and turn it into something I can ship — formatters, image generators, publishing pipelines.
Closing Thoughts
One paragraph on what I’d tell someone starting from scratch — pick one skill, write one hook, then grow the system as friction shows up.