Plan the job
before you
run the agent.
Turn a complex job into clear execution instructions. Define the outcome, set boundaries, plan for failure — then hand the runbook to your agent.
No account. No API calls. Your draft stays on your device.
Define the outcome
A polished public-facing product, with existing features preserved.
Bound the work
Improve copy and user flows. Preserve routes, architecture, and data.
Set approval gates
Implement independently. Return before publishing.
Plan for recovery
Try one reasonable alternative. Keep completed work.
Make done testable
Primary flows work. No placeholders. Changes validated.
Give the job a clear boundary.
“Improve my website.”
The outcome, permissions, and stopping point are left open.
Objective + Context + Scope + Constraints + Autonomy + Approval + Recovery + Done
Decide what good execution looks like before it begins.
Choose the job. Make it yours.
Build a Product Website
Turn a defined product concept into a usable, responsive website with validated primary user flows.
Add a Feature to an Existing Application
Implement the requested feature in the existing application while preserving its architecture and working behavior.
Productize an Existing Prototype
Transform the existing prototype into a coherent public-facing product while preserving working routes and functionality.
Audit and Improve an Existing Website
Identify and fix the most consequential usability and product issues on the existing website, with evidence of improvement.
Designed for capable AI agents, including GPT-6 Astra.
Customize here. Copy the runbook. Execute in your own agent environment.