Or does it only apply upon realization of the solution? The player introduced the solution of the dragon hoard. It’s only a possible solution, but it is a solution.
Yes, this. There's a huge gulf, and tremendously rich gameplay between knowing a solution and making it happen, such that it offers the players a lot of agency and buy-in with very little overhead. The only real requirement is for the GM to let go of 'their precious', relax and participate in the discovery of the world.
What about play that goes more like the below? I’m curious how you’d classify it.
Player: I need cash.
GM: You’ve heard that there’s a dragon in the hills who has a massive treasure hoard.
Player: Okay...good to know. Not sure if I need cash that bad. Have we heard about any other opportubities?
GM: Make a (relevant skill) roll.
Player: Okay....(rolls)
GM (checks results): You’ve also heard that there are bandits in the forest and they’ve been waylaying merchants. There is a bounty being offered for their leader. And, you’ve heard that the northern outpost has been having issues with orcs, and they need folks to help hunt the creatures down. The captain there will pay for help.
Here the GM introduces the possible solution...or solutions, in this case...based on the player’s indication of what the character wants. How would you categorize this example? Agency dead, laughable, alive, limited?
Ha! I may often be grumpy and abrasive, but I really prefer not to classify specific instances of other people's play
So your example features interesting points of discussion. And points to reflect on as well, which is always a good habit to be in.
This exchange:
Player: I need cash.
GM: You’ve heard that there’s a dragon in the hills who has a massive treasure hoard.
Is just the kind of thing I might say in a game. Not because I have a lot of prep done with a dragon
which I want to show off, but to ask the character the question... "just how badly do you need cash?".
And in fact your character answers exactly that question. We learn something about the character... not their skills or their stats, but something about their personality - cautious, or maybe lacking confidence. If I'm the GM, that's an interesting exchange... it makes me want to see this character in situations which put pressure on their caution or confidence.
Okay, so we get to this:
Player: ...Have we heard about any other opportubities?
GM: Make a (relevant skill) roll.
Player: Okay....(rolls)
GM (checks results):
Does the player know why they're rolling? Do they know what number equals success? Do they know what hitting that number means for the fictional outcome? Does failure do anything interesting?
My instinct from your example is the answers are: not precisely, no, no, no. Of course, that may be a doing it an injustice. But that kind of action resolution exchange sends up a lot of smoke signals of GM control and player passivity.
Then the list of 'jobs' looks a bit generic and scripted. I'd be looking for the player to be generating a lead that interested them and then engaging the mechanics to see how much of what they want actually happens.
So, instead, what would you make of this?
Player: Have we heard of any other opportunities?
GM: Who are you asking?
Player: Well, I know this vagabond Harskold who kicks round the streets. He usually knows plenty.
GM: True enough, he does. Not the most reputable sort, though. Sometimes keeps the wrong company.
Player: Yeah. Well maybe that's the sort of work I'm looking for. I'll go look for him.
GM: Okay, Streetwise 16+. If you make it you find him somewhere quiet and comfortable. If you fail - he's going to be somewhere compromised or uncomfortable. You still want to roll?
I would say the player has created a new character in the streets, the GM used a 'Yes.. and...' to flesh out that character with a bit of an edge which let the player know what sort of work they're likely to get. And then the stakes of the roll have been set - not absolutely nailed down, but enough to work with so that we know we're going to get a new situation from the roll; either negotiation over work, or maybe Harskold in the stocks or a gibbet, or a cell and we each get some new ideas for conflict, risk and reward.