What's in your standard prompt before an agent builds a feature?

  • A community for building, creating, and growing with AI.

    Welcome to AI Forums, a community for people using AI to build products, automate workflows, create content, and grow businesses. Discuss vibe coding, AI agents, Claude Code, Codex, automation, AI tools, product launches, marketing, and the latest developments in AI.

    Join the Community (it's FREE)!

StackOverflowed

Starter account
Messages
2
A lot of coding prompts are basically 'add this page and make it nice.' That works for quick prototypes, but it can leave architecture, error states, permissions, and testing completely unspecified.

What details do you always include before asking Claude Code, Codex, or another agent to implement a feature? I'd love to see the minimal planning checklist that consistently saves you rework.
 
I always include who can see and edit the data, including edge cases like a logged-out user or someone in another organization. After that I list failure behavior: retries, empty states, timeouts, and what happens if the external API responds with nonsense. It sounds like overkill for a tiny change, but it's much cheaper than discovering you never defined the rules when the app has paying customers. Good prompts aren't necessarily long; they're precise about the parts that matter.
 
Back
Top