I asked Quinn to try out her [Vulpine Agility] trait. The Abattoir’s basement served as a good testing area. Quinn did some basic jumps, runs, and other maneuvers in the big room. She was quick and light on her feet, and like a cat, she could turn herself easily while in mid-air. Then she jumped and bounced off the stone spikes a few times. I added some ledges and steps around the area to make parkour courses for her to run. [Vulpine Agility] was a passive trait, meaning Quinn didn’t need to activate it or even think about it. She ran my basic courses and obstacles as if she had ten more years of experience.

  After an hour of running and jumping, Quinn took a break and rested while I reset the basement.

  I was curious, though. “Quinn, do you need to eat?”

  My dungeon didn’t have a functional ecosystem.

  “I’m not hungry,” Quinn answered. She put her hand on her stomach, “I don’t think I even need to eat to survive. I can feel that you are sustaining me directly, maybe with mana?”

  That seemed plausible. When I had gotten my quest to get a companion, I had learned that I was previously leaking a lot of mana internally. The leak had been fixed, but the cause had never been something I thought too much about. Now that I thought about it, the mana that I had been losing was probably meant for sustaining my original companion. The changes bumped my mana regeneration sealed the open wound.

  But now that I had Quinn, my mana regeneration was still the same as before. That begged the question: did my mana generation rate increase by the same amount or was I sustaining Quinn another way?

  Perhaps this was the red box’s reward.

  I reeled back my ruminations and built a small stand with some mushrooms on it. I asked Quinn to try eating them. She did–she noted they tasted bland and were really chewy–but she didn’t suffer any indigestion or poisoning. Nothing in my status, particularly mana regeneration, changed when she ate the food, so dungeon food apparently didn’t impact upkeep or provide any boosts.

  That moved us to the next topic–sleep.

  Even being winded after her aerobics, Quinn didn’t feel sleepy. However, she still knew the answer to the question; she did need sleep. That could be potentially problematic, but hopefully the Holding system would allow us to maintain a workable sleep schedule.

  “Quinn, I have a mission for you.” I texted her tablet. “I want you to explore my dungeon and think of ideas for how to improve it.”

  “Humm, sure!” My “father senses,” dulled as they were, sensed her cheerfulness was mainly to cover her lack of confidence.

  “I’ll give you some details of all the rooms, traps, and monsters before you go. Any ideas you have will surely be helpful!” Aping a dwarf game I had tried and failed miserably at playing, I used ASCII-type characters and letters to make a very simplistic map that appeared on her tablet.

  I spent a tiny bit of mana and modified the dungeon to make it accessible to Quinn–a ladder for climbing up from the basement to the bridge, removing the slab that covered the secret stairs, and, although I didn’t know if they would trigger for her, deactivating the traps.

  While she went and reconnoitered the dungeon, I went to work making a new expansion. I wanted to make Quinn a living space and I felt a new dungeon room was warranted too. Actually, I wanted to build a castle (princesses should be in castles, of course), but building a castle above ground would just invite its inevitable destruction by the locals. On the other hand, carving out a large open space underground to put a castle into it would just be a waste of resources. Then again, it didn’t need to be a castle, it just had to look like one to the delvers.

  The next section of my dungeon would be the sewers, because of course it would be.

  First, I use landscape to make a half-dozen square metal sluice grates in the floor of the basement. They were arranged haphazardly in between the stone spikes, but using Landscape, I was confident they were securely in place and couldn’t be pried off.

  That is, all but one, which I created with some hinges so it could be swung open. I added a sturdy iron lock. Then, I put a key on a necklace on the now-crownless Skeleton Knight. Using Trap Creation, I paired the two so that the key had to be used to open the grate.

  The grates topped holes that went down about six feet and which opened to another room. I made it as small as possible while still having all six holes empty into it, and added a dozen carved stone columns that went from the floor to the ceiling. This made the room slightly cramped; someone could move around but they couldn’t go in a straight line, having to meander around the columns. The floor wasn’t flat, but had shallow channels running around, just as if it was a sewer with running water (that would come later).

  At one end, I continued to cut a winding tunnel that slanted just a little bit. I absorbed the rock so that it looked like it was originally natural, but had been expanded in some places by hand. The path was difficult and varied–very narrow or short in some spots, columns of rock splitting the path, and sharp angled rocks jutting out in the middle in other spots.

  In total, the sewer tunnel went for about 150 feet. At the end, I dug another hole that went about twenty feet down.

  About ten feet before the end of the tunnel, I carved a side passage that turned into stairs and then a small platform where there was a door. This would serve as the entrance to the new room (or rooms, I wasn’t sure, yet).

  Now, it was time to add a core element to any sewer–the running water.

  However, this would be fairly tricky to make work.

  Well, the easy option would be to make a very, very large cistern higher up that would feed water into the sewer. If the cistern was part of a Holding, I could probably figure out a way to get the water to reset each time the holding was instanced. This was just a variation of what I did for the grotto room.

  But it didn’t seem very economical to me. Also, I wanted to push my skills even further.

  This tale has been unlawfully lifted from Royal Road; report any instances of this story if found elsewhere.

  So at the top of the incline, I carved out some holes at the bottom of the wall. Then I set the rock inside as part of a trap. Even though the sewer was not designated as part of the Abattoir, I set the trap’s trigger condition as when that Holding instanced. It worked.

  The trap was actually very simple: when the trap was triggered, the rock sections would use Landscape to produce water. The water would flow naturally down the gradual decline of the tunnel, using the channels I had cut in the floor, and then waterfall at the end into the pit. The rock at the end of the pit was set to Absorb the water when the water in the pit reached a certain level (about four feet).

  That was a good base, but I wanted it to be a bit more.

  I used Landscape and Trap Creation together to modify the water that would be produced. First, I made it so the color of the water would be blood red. Then I used Trap Creation to make the water extra slippery–not just a “water is wet”-type of slippery, but to actually use the mechanics of a trap to give that slipperiness some extra magical oomph and effectiveness.

  I had neglected the Landscape for some time (to be fair to myself, Trap Creation was awesome), but that didn’t need to be the case any more. Landscape and Trap Creation had a lot in the way of synergy. However, the way I was creating dynamic environments was janky, at best. I really should push Landscape more.

  Hm. Why not now?

  I deleted the slip-n-slide sewer trap entirely.

  Starting with the water-stones at the top and proceeding slowly, I used only Landscape to try to make the water hazard again. It took several tries to get it right as I had to create an initial layer of intent over the whole thing, then fill in the remaining parts. However, once it all clicked into place, the water rushed out of the starter blocks and flowed as intended. In addition, the sewer’s stone floor and the water had an extra slipperiness to it just as I intended.

  Trap Creation created traps, and while I abused that function in many ways, it was still creating something that was meant to be interacted with, and beatable. Here, Landscape created something embedded within the environment and terrain itself. This was a small but significant difference and the system agreed.

  Also, as part of the terrain, I didn’t need to embed it with any signals regarding the Holding for it to work (it was already gushing water now). This would avoid a lot of problems when I had more than one Holding.

  I returned my attention back to the end of the tunnel.

  Here, I expanded my demesne to a large square area–a cube 50 feet on each side–and absorbed everything within. Then, I cut more at the top to create a domed ceiling.

  In the very center of the dome I made a mass of moss and pushed it together as tight as I could. Then I pushed more and more moss into the mass until it was even denser. For color, I chose a soft yellow similar to the type of light old incandescent bulbs produced. The effect was immediate: the ball of moss shined incredibly brightly and bathed the empty room in light.

  I delicately carved into the walls, cutting lines of stone so that the walls looked like they had been constructed. I added four pilasters on each of the side walls and would moss over them as if they had grown like ivy. I chose blue for their color, and dialed their brightness almost all-the-way down so they barely gave off any light.

  In between the two central pilasters, I carved an arched entryway that recessed only about a foot into the stone.

  Here, I paused.

  I intended one of the functions of this room to be a portal room (aesthetically, it was the courtyard in front of the castle). However, I was hesitant–very hesitant–to do anything with that skill after what had happened with Quinn. The black mana had a will of its own and was also connected to that disgusting skill that the goblins had. Tenebrous Portal was an incredibly useful skill, but it was not completely under my control. As the kid with the dwarves said, “Don’t touch it, it’s EVIL!”

  But . . .

  Portals were just so damn useful.

  So, I filled the archway on the right side with a portal and covered it with a big, heavy, wooden door. Behind the door, I put in a portal.

  Nothing unusual happened, which was a bit of a relief.

  For the front entrance, I made an even larger entryway with a stone archway as nicely crafted as I could make it. I then installed a set of double doors, much larger than the side door. Behind the doors, I filled the entire oversized entryway with a second portal. This would be the main entrance to the interior of the (underground) castle.

  Structurally, it would be obvious that the entryway to the castle would lead to a room on the other side of the wall. But with a portal, that didn’t have to be the case; I could put the next room anywhere.

  I sent out a dozen thin tendrils of demesne in every direction (except directly down, which I wasn’t too concerned with), twisting and twirling but always expanding outwards. When any of the tendrils hit open air, I stopped going in that direction and expanded it at a slightly different angle. I found it relatively easy to maintain all of the threads going out at the same time. This was probably due to my higher level of Demesne.

  I ended up mapping a good amount of three-dimensional space.

  I learned that the “Gorge” in my name might not be random. My core room–the original one–was actually very close to a very large and very steep escarpment. In fact, the caved-in section was directly exposed to the outside on the other side. Anyways, the escarpment wasn’t uniform–some places were just steeply sloped, others were sheer cliffs, and there were some ledges too.

  The cliff gave me some ideas that fit well with my castle theme. So I deleted all of the other threads and expanded towards one particularly large sheer cliff section.

  I did some basic calculations and picked a spot in the interior to make the next room. This one I also gave a large footprint, but rectangular in shape (60 feet wide by 40 feet).I also made it about thirty feet high.

  I cleaned it out and made another portal entry way that matched the large one in the courtyard room and installed the portal’s twin.

  Now, with three portals waiting for connections, I could make some more transitions.

  But first, I quickly checked on Quinn to see how she was doing and–

  –she was waiting close to my core. She looked bored and a bit peeved.

  Oops. I think I had been ignoring her for a little while.

  I quickly typed out a message, “I’m sorry! I was a little busy with the new expansion.”

  “It’s fine,” she dragged out her answer in a way that I knew it was the opposite.

  “So, what are your thoughts?”

  “I have some questions.”