I'll be walking through the woods, enjoying the quiet, when I suddenly think of a door that opens only when someone tells it a convincing lie. Or I'll be sitting at a cafe, drinking an espresso, when a puzzle involving four statues, a dead king, and a missing crown pops into my head.
Naturally, that's when I have absolutely nothing useful to write on.
I've scribbled ideas on napkins and stopped in the middle of walks to type a few words into my phone. I've also written notes that made perfect sense at the time, only to find them weeks later and wonder what on earth I was thinking.
Three bells. Dead priest. Don't ring the second one.
Thanks, past me. Very helpful.
Some of those ideas reach the table. Others sit around for years. When my players finally approach a ruined temple and I need a puzzle, I often can't remember a single useful one.
So I open Reddit, dig through old discussions, and look for something I can adapt for tonight's session. Over time, those searches and my own ideas have left me with quite a pile of notes.
I've gathered some of them in The Puzzlekeeper's Grimoire, a free collection of encounters for D&D 5e. First, here's what I've learned about making puzzles work at the table.
Start with the discovery
There's a moment I love: one player notices a detail, someone else remembers an earlier scene, and suddenly everyone starts talking at once. That's usually when I know the puzzle worked.
An elaborate mechanism can fall flat. A simple connection can keep players talking long after the session ends.
Before I write a riddle or a puzzle room, I decide what I want the players to realize. Perhaps an old legend explains the carvings. Perhaps the room shows what happened to the people who built it. Once I know the discovery, I can work backward and give the party enough evidence to reach it.
Give them the clues before they need them
Imagine an old woman telling the party about four brothers who founded a kingdom. She gives their names and birth order. Later, the characters see the brothers in a chapel mural. Several sessions after that, they find four statues blocking a royal crypt.
The inscription asks them to place the brothers in order. They've already encountered the answer. Now they have to connect it to the statues.
I like this because it gives an earlier conversation a purpose. The painting they stopped to examine becomes useful.
Be reasonable about memory, though. A name mentioned once, eight sessions ago, is asking a lot. Bring important details back in another form. Let a successful History check recall the legend, or an Investigation check reveal how the mural connects to the mechanism.
Leave a deeper answer for curious players
Once the party arranges the statues, the door opens. They can continue the adventure.
But one statue is missing a finger. The same detail appears in the chapel mural. A player who examines that hand finds a concealed mechanism and opens a compartment containing a relic or an old letter.
That's the kind of discovery I love. Everyone gets through the door, and the players who look closer find something extra.
You can do this with riddles, too. An inscription has a straightforward answer, while the first letters of its lines spell a second message. A device has one sequence that opens the passage and another that disables the guardians beyond it.
The extra reward can be small. A map or a clue about a future adventure is enough if the discovery feels earned.
Let the puzzle respond
A player moves a statue or touches a rune. You say, "Nothing happens." They try another idea. Nothing happens again.
If every wrong attempt gets the same response, the players have very little information to work with.
Show them something. A rune lights up when a statue faces the right direction. A misplaced piece makes the mechanism grind. A flame changes color when they use the wrong element.
Decide how the mechanism responds before play so its feedback stays consistent. An unsuccessful attempt can still teach the group a rule they can use next time.
And pay attention when they try something you hadn't considered. Their solution might be better than yours.
Remember that your players have swords
And spells. Lots of spells.
You can spend hours building a challenge around a locked door, only for the wizard to cast knock. It can open a mundane lock or suppress arcane lock for ten minutes. The barbarian may try removing the wall around the door instead.
Honestly, I find that pretty funny.
Check what the ability actually does, then let it work when it applies. The characters have those tools for a reason.
The world should still make sense. A reinforced vault won't collapse because someone hit it with a chair. Forcing a mechanism can damage what's inside or alert someone nearby. Establish those consequences from the situation.
I don't like protecting a puzzle simply because I spent a long time preparing it. Sometimes the best moment of the evening comes from a solution I never considered.
Prepare puzzle hints before the session
A puzzle that seems obvious while you're writing it can become surprisingly difficult when other people try to solve it. You already know the answer. Your players are juggling clues, competing theories, and everything else happening in the adventure.
I prepare a few hints. The first points to an overlooked detail. The next explains a connection. The last gives the group enough information to make progress.
You can deliver them through an NPC or an appropriate ability check. If the clue is essential to the only route forward, give the party another way to find it when a roll fails.
I also avoid putting the whole adventure behind one mandatory riddle. An optional treasure room can hold a difficult puzzle. For the only route forward, I provide another way through, even if it costs the party a spell slot or alerts the dungeon's guards.
Watch who's still involved, too. If one player enjoys decoding the inscription and another is losing interest, give the second player something useful to do, such as searching the room for another clue.
Building puzzles from La Notte Eterna lore
La Notte Eterna 5e takes place in Neir, a world that has lived without sunlight for more than a century. The gods are fighting a Celestial War, and creatures once confined underground have claimed territory on the surface.
For a Dungeon Master, the changes leave plenty of places where an old mechanism can reveal something about the world.
Take the temples of Laon, the imprisoned god of the Sun. Many have been abandoned or corrupted since his capture. Their carvings still show an age when the Sun rose every morning.
Imagine the party finding a mechanism that seems to require sunlight. That gives them an immediate problem, but Laon's story holds a clue: he sacrificed part of his brilliance so the Moon could continue to shine.
I'd build an encounter around redirecting moonlight toward the mechanism. The carvings could connect the Sun and Moon, giving the players a reason to try it. A closer reading might reveal another function the priests concealed.
That mechanism is a new encounter idea built from the setting's lore. Solving it lets the players learn something about Laon through their own actions.
Lanie, the fallen goddess of magic, offers another approach. Her temples and artifacts survived the collapse of her cult. So did the Dream Dragons she created as messengers, which can guide and test mortals through dreams.
Imagine a village where the same child drops a basket of apples every time the party crosses the market. The villagers repeat conversations. An old man recognizes a character he's never met.
As those details accumulate, the players can realize they're dreaming. The dragon's disguised identity becomes another mystery to solve.
In the Domain of Doomkhan, Dream Dragons sometimes test people seeking knowledge in its immense library. A riddle could point the party toward the right text, giving them a reason to explore the collection.
Then there's Maskarr, the god of rogues. Beneath the Cat's Palace lies a labyrinth of mechanical traps and constructs guarding his treasure. I can easily imagine a chest that opens only when someone tries to steal its key. That sounds like the sort of trick he'd enjoy.
These places give the challenge a reason to exist. When the players understand how it works, they also learn something about whoever built it.
Download The Puzzlekeeper's Grimoire
After years of collecting ideas, losing them, and searching for replacements, I finally organized some of that material into The Puzzlekeeper's Grimoire.
It's a free, 76-page PDF in US Letter format, with 55 riddles, puzzles, traps, and challenges for D&D 5e. Eleven encounters are inspired by La Notte Eterna. Others adapt ideas from discussions, articles, and community contributions, alongside material I've developed myself.
The entries include a scene to present, a solution, staged hints, an alternative approach, and an optional discovery. You'll also find diagrams and printable handouts where something visual helps.
The source notes credit creators where their authorship could be confirmed, and the directory links to the original material. I have no interest in taking credit for someone else's work. Visit those posts and see what else their creators have shared.
Read through an encounter before running it, then adapt it to your group. Change the rewards, replace the creatures, or turn a small riddle into a larger scene. You know your players better than I do.
As for me, there's still a pile of notes on my phone to sort through. I'm pretty sure a cafe napkin somewhere has a brilliant puzzle written on it.
At least, I remember thinking it was brilliant.
I just wish I could remember what the puzzle was.
Share this article:
