WorkshopsWERKBANK

Workbench

Hands-on. Real agent projects, step by step.

From empty repo to tool-loop in one hour - a guided walkthrough.

One terminal script, one model, one answer - the smallest building block of an agent, built straight against the API.

The one-shot call becomes a conversation - the history is just an array that rides along on every request.

Your agent gets an identity - your answers become a system prompt that lives in its own file.

Now it becomes an agent: tools that fit its purpose and the loop that keeps working until the task is done.

Limits make the agent fit for daily use: confirmation before risky actions, an iteration cap, and a list of hard nos.

A memory made of one file: your agent remembers what matters - and still knows it after a restart.

Your agent runs while you are away: one job, one schedule, one log - automation without frills.

When the conversation outgrows the window: older turns get summarized - the thread survives, the cost drops.

The finale: a small model handles the tool calls while the big one thinks - a division of labor you can measure.