Connecting
Learn/L14 · AI Coding Workflow

How to prompt AI coding agents

Direct the agent like a junior engineer: architecture first, scoped tasks, review every diff.

Difficulty 3/5

Plain English

Don't ask 'build me the app'. Ask: 'here is the schema, the existing files, the convention, the constraint. Add one feature. Show me the plan first.' Approve the plan. Then ask for the implementation. Then read the diff line by line.

Analogy

Treat the agent like a contractor: blueprints first, walkthroughs after each phase, sign off before the next.

Jessica's wedding portal

Bad: 'make the calendar work'. Good: 'Add a month calendar view at /events/:id/calendar. Read tasks via existing useQuery. One file per component. Show plan first.'

Agentify CRM

Bad: 'add Telegram support'. Good: 'Add a Telegram channel reader. Schema already has agentify_conversations with channel enum. Add a new edge function that receives webhook, parses, writes a message row. Plan first.'

Go deeper

Architecture → small task → review.

Common mistakes

  • Letting the agent invent architecture.
  • Accepting code you can't read.
  • Letting the agent silence errors instead of fixing them.

AI-agent trap

The biggest one: blind trust. Always read what it produced.
Quick check
What should you ask for before the agent writes code?

Summary

Always provide: relevant files, existing conventions, the data shape, what to NOT touch. Ask for a plan before code. Reject 'any' and '@ts-ignore'. Demand small commits.

Confidence: