Agent 04 of 8
The agent that decides what is actually worth doing
Scores every gap and ranks them, highest impact first.
A list of two hundred keyword gaps is not a plan, it is a way of feeling busy. The hard part of SEO has never been finding opportunities; it is knowing which six to do this quarter.
This agent scores what the research surfaced and puts it in order. The output is a ranked list where the top of it is the argument — this one, before those, for these reasons — rather than a spreadsheet you are left to prioritise yourself.
It is deliberately the step between research and writing. Nothing gets drafted until something has decided it is worth drafting.
What it is handed
- The demand map and keyword sets from agent 03
- Your existing pages and their performance
What it produces
- A ranked list of opportunities, highest impact first
What it does not do
It is a scoring engine, not a language model — the ranking is computed from the signals, not argued by a model. That makes it consistent and repeatable, and it means the reasoning is arithmetic you can check rather than prose you have to trust.