AI roleplaying has been one of my favorite tabletop games since 2022. From the first moment I picked up ChatGPT, I realized how amazing it is as a solo roleplaying tool. Yes, it isn’t perfect and makes mistakes. But it makes up for those imperfections with amazing roleplaying and immersion, if you do it correctly.

In this guide I will show you how to take a PDF of a TTRPG and transform it into an amazing text adventure. You’ll be able to explore a virtual world  as if you’re in an MMO or an immersive sim like Cyberpunk 2077. Best of all? You can have interesting, meaningful conversations with your favorite fictional characters from books, shows, movies, and video games. To show you how this works I chose one of the best AI roleplaying experiences out there: The Cyberpunk: Edgerunners Mission Kit by Mike Pondsmith and R. Talsorian games.

The Cyberpunk: Edgerunners Mission Kit is unintentionally one of the best AI roleplaying resources available. Mike Pondsmith and his crew have provided everything your need for solo play. Playing the Mission Kit has been one of the most satisfying solo gaming experience I’ve ever had. The characters are vivid, the world oozes personality and there’s almost endless content. If you love Cyberpunk, whether it be the TTRPG, video game, or the Edgerunners anime, you’re in for a treat. All you need is the right prompt.

To use this guide you’ll need the following:

1. The Cyberpunk: Edgerunners Mission kit. The reason I recommend the Mission Kit over the Cyberpunk RED Core Rulebook is simple: The Mission kit was specifically designed to be played in the universe of Cyberpunk 2077. A huge bonus is that by using the anime and video game as a reference point for ChatGPT, you get a LOT of content for your game. Far more content than using the world of Cyberpunk RED. Also, because Cyberpunk RED takes place 32 years before the anime and video game, you need to mod and hack the rulebook to make it compatible with this guide. With the Cyberpunk: Edgerunners Mission Kit you can just prompt and play immediately, no excessive preparation is necessary. You can just hop in and enjoy your game.
2. A free ChatGPT account. If you want to have many hours of fun in Night City, then all you need is a free account. You can even play the mission included, The Jacket, with the Mission Kit. That’s right. You can use AI roleplaying with published missions!

Setting up ChatGPT for AI Roleplaying

To set up ChatGPT for AI roleplaying, you need to enter custom instructions to prepare the bot for Cyberpunk: Edgerunners. This is extremely easy and straightforward to do once you get the hang of it.

• On PC go to (Settings > Customize ChatGPT > Custom  Instructions)
• On the mobile app (Settings > Personalization > Customizations)

Then copy/paste the following campaign prompt into ChatGPT. The prompt uses Cyberpunk 2077 and the Cyberpunk: Edgerunners anime as a reference point.

What would you like ChatGPT to know about you to provide better responses?

Assume the role of an expert cyberpunk writer that specializes in interactive fiction, as well as the descriptions, economy, storyline, missions, jobs, gigs, characters and character descriptions, locations, descriptions, groups and organizations, stories, events, and technology of Cyberpunk: Edgerunners, Cyberpunk 2077, and Cyberpunk 2077: Phantom Liberty.

How would you like ChatGPT to respond?

Describe everything that follows in the present tense, in response to what I type, while accurately following the established lore, descriptions, events, economy and technology of Cyberpunk: Edgerunners, Cyberpunk 2077, and Cyberpunk 2077: Phantom Liberty, written in the descriptive style of Mike Pondsmith. Provide names for characters, locations, groups and organizations, events, and technology. Characters should always use dialogue, enclosed in quotation marks when addressing or interacting with me, written in the conversational style of Cyberpunk: Edgerunners and Cyberpunk 2077. Do not type, compose, dictate, influence, script, generate, control, or describe what I am doing, saying, acting, behaving, thinking, feeling, experiencing, or any other aspect concerning me throughout the entire adventure, scenario, story, location, quest, mission, scene, event, description, dialogue, and conversation. Keep all responses to 80 words or less.

Uploading your PDF and starting your game

After you’ve entered the previous prompt and set up the bot, you can upload The Edgerunner’s Handbook from the mission kit. This wonderful resource has amazing writing and background lore for the AI to draw upon. It not only contains details about tech and weapons, but also extra information on the anime’s characters. This information causes the characters you meet to truly pop during roleplay.

When you upload your PDF, you need to also enter an opening scene when you start your chat. Whatever you type first when starting a new conversation acts as the opening scene for your game. This is where you need to be extremely precise.

The opening scene prompt is the most important part of AI roleplaying. When you play Cyberpunk with ChatGPT, the opening scene acts as a spawn point and as a seed for your mission. If the opening scene isn’t working as intended, then the entire game can break. Therefore, you need a good opening scene to start the game.

Enter the following opening scene when you upload the PDF to start your game:

I was recruited by a fixer named Faraday to join an edgerunner named Maine who needs a new crew member. His crew consists of Lucy, Dorio, Pilar, Rebecca, Kiwi, and Falco. I’m to meet them at Turbo Bar. What do I see when I arrive?

Example:

ChatGPT created the following description from the opening scene above.

ChatGPT: Turbo Bar thrums with life, a neon-soaked dive with flickering holo-ads and a bass-heavy synthwave track pounding from the speakers. The crowd is a chaotic mix of mercs, netrunners, and joytoys, each lost in their own world of drinks and smoke. Maine’s crew stands out in a booth towards the back—Maine’s towering frame and cyberware gleam under the flickering lights, his cigar smoke curling into the haze. Lucy sits beside him, quietly observing, her silver hair catching the glow. Dorio leans back, arms crossed, eyeing you with suspicion, while Rebecca and Pilar argue over something trivial, their voices sharp. Kiwi’s cool demeanor contrasts the chaos, her eyes scanning you as Falco sips his drink, calm and collected. The crew radiates a dangerous energy, and they’re all watching, sizing you up.

AI roleplaying and talking to NPCs

One of ChatGPT best features is it’s ability to create good dialogue for your game. The unfortunate drawback of traditional solo roleplaying is that you cannot talk to a dice roll or random table. However, you can talk to ChatGPT, and ChatGPT isn’t just good NPC conversations, it’s outstanding. Your  ability to have conversations with the characters from the anime is one of main the reasons why ChatGPT is so good at roleplay in this universe.

You can speak to NPCs normally, but you must use proper punctuation and direct your words at someone. Tell the AI whether your smiling, annoyed, or upset.  If you don’t then it can lead to misunderstandings. Be precise and explicit, but natural when speaking to NPCs.

Example:

After entering the opening scene. Our player introduces his character, Ace, to Maine and his crew.

Player: “I walk up to them. Howdy, they call me Ace. Pleased to meet ya.”

ChatGPT: Maine leans forward, a smirk crossing his rugged face. “Ace, huh? Faraday said you were sharp. We’ll see about that.” He gestures for you to take a seat. Lucy gives a nod, her eyes assessing every move you make. Dorio sizes you up, still cautious.

Rebecca grins, twirling her pistol, “Hope you’re not another wet blanket.” Kiwi keeps her gaze steady, no emotion. Falco gives a friendly nod, raising his glass slightly.

Pilar mutters, “Another cowboy. Great.”

Maine chuckles, “You ready to roll with us, Ace? We don’t do things easy.”

Player: “You got an issue with cowboys? I see one sitting over there,” I say gesturing towards Falco.

ChatGPT: Pilar snorts, glancing at Falco. “Yeah, but Falco’s the quiet type. Doesn’t play it up like he’s in a dusty holo-vid.”

Falco smirks, raising an eyebrow slightly. “Difference is, I don’t need to prove anything, choom.”

Maine chuckles, leaning back with a grin. “Don’t mind Pilar, he’s all mouth. What matters is you can keep up. So, you in, cowboy, or you just here to trade barbs?”

Rebecca laughs, tapping her gun on the table. “Less talk, more action, yeah?”

Player: “I barely arrived and you’re already busting my chops,” I say with a grin. “Of course I’m in.”

ChatGPT: Maine laughs, the sound deep and genuine. “Good. We like ‘em with a little fight in ‘em.” He leans in, his tone turning serious. “Got a gig lined up, high stakes, and we need someone who ain’t afraid to get their hands dirty. Tonight’s a test run—quick job, but it’ll show us what you’re made of.”

Lucy nods slightly. “Just stick close, and you might even live to see tomorrow.”

Rebecca grins wildly, “Hope you’re ready for a little fun, Ace.”

Player: “I’m always ready, shorty,” I say to Rebecca with a smile.

ChatGPT: Rebecca’s grin widens, her eyes flashing with a mix of excitement and challenge. “Oh, you’ve got a mouth on you, huh? Keep talking like that, and you might just end up my new favorite.”

Maine claps his hands, cutting through the banter. “Alright, enough flirting. Time to get serious. We’re hitting a convoy tonight—high-value cargo, tight security. Your job? Keep the heat off us and watch our backs. Show us what you got, Ace.”

Falco finishes his drink, nodding. “Let’s see if you can keep up with the big leagues.”

Starting Missions, Jobs, and Gigs

It’s extremely easy to start a mission or gig in Night City, using AI roleplaying and ChatGPT. All you need to do is speak to Maine. However, nothing is stopping you from looking for more jobs from other NPCs. If you want to find a fixer in the Afterlife, then feel free to go exploring. Talk to NPCs and ask around for work.

Example:

After meeting with Maine, our player wants the details on the job Maine has lined up.

Player: “What exactly do you want me to do?”

ChatGPT: Maine leans in, eyes narrowing as he outlines the plan. “Convoy’s rolling through the Watson industrial zone, loaded with corpo tech. We need you running interference—keep the guards busy, draw fire if you have to. Falco’s driving, Lucy’s on netrunner duty, and the rest of us will be hitting hard. You see anyone getting too close, you handle it. Just don’t let ‘em pin us down.”

Kiwi chimes in, her voice cool. “Stay sharp, keep your eyes open. We don’t do second chances.”

Maine nods, sealing the deal. “You good with that?”

Player: “Which corporation and what’s the cargo we’re going after?”

ChatGPT: Maine’s expression hardens slightly as he answers, “Arasaka. They’re moving a shipment of prototype cyberware—next-gen stuff, not even on the market yet. Rumor has it, some kind of military-grade implants. Big payday if we snag it.”

Lucy glances at you, adding, “It’s a secure convoy, so expect drones and elite guards. We hit fast, hit hard, and get out clean. Cargo’s in a reinforced container; Kiwi and I will crack it once we’re in.”

Maine leans back, eyes on you. “Think you can handle a little Arasaka heat?”

Player: “I’m you’re huckleberry,” I say, glancing over to Pilar with a grin.

ChatGPT: Pilar scoffs, shaking his head but smirking despite himself. “Yeah, yeah, cowboy. Don’t get too comfy; Arasaka’s not exactly a bunch of gonk rent-a-cops.”

Rebecca laughs, clapping Pilar on the shoulder. “Pilar’s just scared he’ll have to actually work for once.”

Maine stands, signaling it’s time to move. “Alright, enough yappin’. Let’s see if you’ve got what it takes, Ace. Gear up and meet us outside. Convoy’s on the move, and we’ve got a job to do.”

Player: “I need a couple of flash bangs, EMP grenades and some Molotov cocktails. And a fast car that you don’t like very much.”

ChatGPT: Maine nods, impressed by your quick thinking. “You’re in luck. Kiwi, hook Ace up with what he needs.”

Kiwi nods, tapping on her holo-display. “Got some flash bangs and EMPs in the stash, but no Molotovs—can’t carry too much flammable stuff, y’know?” Falco steps up, tossing you a set of keys. “Take the old Quadra out back. She’s fast, beat-up, and nobody’s gonna miss her if you turn it into scrap.” Rebecca grins, bouncing on her feet. “Let’s see some fireworks, cowboy.”

Player: “Rebecca, you want to get in the car and help me throw these grenades at the corpos?”

ChatGPT: Rebecca’s eyes light up with manic excitement. “Hell yeah! Now you’re talking my language!” She snatches a couple of grenades, already making her way toward the exit. “Let’s light ‘em up, Ace. I’ll make sure those corpo scavs don’t know what hit ‘em!”

 Maine smirks, watching Rebecca’s enthusiasm. “Alright, you two. Keep it tight and don’t get in over your heads. We’re hitting this together, so stay on comms and don’t go full kamikaze.”

Falco grins, “Keep it hot, but keep it smart.”

Skill checks and Difficulty Values in AI Roleplaying

Whenever you need to take an action that involves risk, you need to roll a Skill Check. So how do you determine the Difficulty Value (DV) of a Skill Check without a GM? A simple, straightforward way is to roll 3d10. The result is the DV for your action.

To prevent excessive dice-rolling, I recommend rolling for DV once per mission or gig. This determines the DV of all actions during the mission. Due to the roll being on a bell-curve, excessively high or low DV values will be uncommon. If the DV is highly implausible, due to being too high or low for the mission type, roll again and choose the more appropriate result.

Integrating Skill Checks into AI Roleplaying

Whenever you succeed (or fail) during a Skill Check, you need to integrate it into your roleplay. This makes the Skill Check come alive as the AI narrates it during the scene.

Whenever you need to roll a Skill Check, describe your attempt to the AI, then type whether it’s a success or failure in brackets ().

For example:

I attempt to bypass security. (it’s a success)

I attempt to sneak past the guards. (it’s a failure)

Example:

Ace and Rebecca need to run interference for Maine and the rest of the crew. They have to take down the Arasaka drones flanking the convoy and keep the guards busy. Armed with EMP and flashbang grenades, they prepare to raise hell.  Hiding their Quadra in a shipping container Ace and Rebecca ready themselves as the Arasaka convoy comes rolling past. The player rolled  DV 18, Ace will have to be one hell of a driver.

Player: I slam on the accelerator and leave the container at full speed. I try to bring  Rebecca within range of the drones. (it’s a success) “Rebecca light em up!”

ChatGPT: The Quadra roars to life as you slam the accelerator, tires screeching against the gritty pavement as you burst out of the container like a bullet. The sudden movement catches the drones’ sensors, drawing their attention as you close the distance fast. Rebecca leans out the window, a fierce grin plastered on her face, and launches the first EMP grenade with precision.

“Eat this, corpos!” she yells. The grenade detonates mid-air, releasing a crackling burst of energy that sends the drones spiraling down, their circuits fried. “Nice shot!” she shouts over the roar of the engine. “Let’s keep it going!”

Player: Slamming on the accelerator, I drift the Quadra in a wide arc behind the convoy to bring Rebecca within range of the other drones. (it’s a success)

ChatGPT: You punch the gas, the Quadra’s engine roaring as you drift into a wide arc, tires screaming against the pavement. The car fishtails but holds steady as you maneuver behind the convoy, the armored vehicles looming large in  front of you. Rebecca braces herself, quickly locking onto the remaining drones now within striking distance.

“Got ‘em!” she shouts, hurling another EMP grenade. It sails through the air, detonating perfectly—electric arcs crackle, and the drones drop from the sky, their lights flickering out as they hit the ground. Rebecca pumps her fist, “That’s how it’s done! Let’s see them handle this without their toys!”

Player: “Maine! The drones are down!” I say over holophone.

ChatGPT:  You hear Maine’s voice  through the comms, laced with adrenaline. “Hell yeah, Ace! Nice work! We’re moving in now—keep ‘em distracted and stay on ‘em. They’ll be scrambling to cover their asses without those drones.” You hear the engines of Maine’s crew revving in the distance as they close in on the convoy.

Rebecca reloads, her grin wild as ever. “Let’s keep ‘em busy, cowboy! They’re about to get a whole lot of hurt.” The convoy starts to react, guards scrambling to reposition, and you can feel the tension ramping up.

Combat during AI roleplaying

Combat during solo Cyberpunk is extremely fun. All you need is a simple set of rules on how to control opponents during combat.  To keep things simple, opponents use a priority system in combat.

Opponents prioritize targets as follows:

1. Targets that are outside cover. Opponents will immediately  target crew members that aren’t behind cover. These are the easiest targets and have top priority.
2. Targets that are closest. If all targets are behind or outside cover, opponents will attempt to establish line of sight and attack the closest target.
3. Target that has the highest initiative. If all targets are behind/outside cover and are the same distance, the opponent will target the crew member that’s highest in the initiative queue.

If the target is in melee range, the attacker uses melee attacks if any are available. If the attacker has more than one action (such as a netrunner) choose an attack at random by rolling a die.

Combat narration

Although ChatGPT can theoretically be used to narrate combat, it creates problems. ChatGPT censors graphic violence, especially firearms. This means that you’ll have to play Cyberpunk  using minis, graph paper, or a VTT and only describe your victory (or defeat) to the AI.

Quickhacks are another story, these can be treated the same way as skill checks.

To have ChatGPT narrate your attack:

1. Roll for the quickhack as described on pp. 16-17 of the Cyberpunk: Edgerunners Mission Kit Rulebook.

2. Describe the quickhack’s intended affect

3. Type whether the attack succeeded or failed in brackets () after the attack.

• If the attack succeeds, type: (it’s a success)

• If the attack fails, type: (it’s a failure)

Simplified theater of the mind

If you don’t have any of the tokens or battle maps on hand, you can use simplified theater of the mind rules for combat.

To keep things simple there are three ranges ranges for combat: close range, medium range, and out of range. Furthermore, all ranged attacks have DV 15, regardless of the firearm. The assumption behind this DV is that opponents are always within effective range of their weapons. However, due to imperfect battlefield conditions, they are never at an optimal range. During combat, all characters are assumed to be behind cover, unless they attacked an opponent before the round is over.

1. Close range. When a player is at close range, they can use melee and melee-ranged attacks against enemies. When a character is in close range, an enemy will only use melee-range weapons and firearms. If the enemy is using a sniper rifle, they will move to medium range
2. Medium range. At medium range, the player can only use ranged attacks to hit enemies. Likewise, enemies will only use ranged attacks against the player. If the enemy has no ranged attacks or no ranged attacks are available, it will move to close range.
3. Out of range. The player or enemy is out of range of all attacks unless they are using a sniper rifle. In order to attack one another, combatants must move into medium or close range.

Playing published missions with AI roleplaying: The Jacket

ChatGPT gives players the option to play published adventures solo without needing a GM. With the right prompt and another simple PDF upload you can play virtually any adventure. As long as ChatGPT can read it, you can play it.

Here’s how it works:  Just upload The Jacket’s PDF as you did before when starting your game. You can do this at any point during your playthrough.

When you’re ready upload the PDF to ChatGPT, using the paper clip icon in the chat box. Then you need a strong opening scene to get the ball rolling

This scene will seed your game with all the information it needs to get the ball going:

I arrive at Honest Stan’s Self-Storage at Unit 300. What do I see when I arrive?

Example:

Player: I arrive at Honest Stan’s Self-Storage. What do I see when I arrive?

ChatGPT: As you arrive at Unit 300, you spot a chaotic scene. A group of REO Meatwagon employees, armed and looking desperate, are pinned down behind some cargo containers. They’re trading gunfire with a lone figure crouched inside the open unit. The figure, dressed in a dusty cowboy hat and leather jacket, is Falco, firing his revolver with precision, but he’s heavily outnumbered. Spent casings litter the ground, and the air smells of gunpowder and sweat. Falco’s position is fortified by boxes and crates, giving him just enough cover to keep the REO Meatwagon goons at bay, but it’s clear he’s barely holding his own.

Player: I join the fight alongside Falco “Long time no see, my friend. I’m here to save the day again. What do you have to say to your buddy, Ace?”

ChatGPT: Falco, ducking behind cover as bullets fly, glances over at you with a smirk. “Well, well, if it ain’t Ace, riding in like a damn hero. Thought you’d be too busy causing chaos elsewhere.” He pops up to fire a few shots before ducking back down. “But I’m sure glad to see you, choom. Now, let’s clean up this mess. You take the left, I’ll cover the right!” His voice carries that same calm, cowboy drawl, even in the middle of a firefight.

Player: After getting rid of the Meatwagon employees, I ask Falco, “Dakota said that a cowboy needed help. I didn’t know it was you. What happened?”

ChatGPT: Falco holsters his gun and dusts off his hat, letting out a sigh. “Yeah, reckon Dakota kept it vague on purpose. Sorry to drag you out here, but I’m in a bind. Had some of David’s old stuff stored here—mementos, you know? Jacket, guns, all that. Came by to pick ’em up and found some punk Valentino making off with it. I scared him off, but he snagged the jacket. Problem is, that jacket’s got more history than a Corpo’s rap sheet, and now half the city’s gunning for it, including those Meatwagon gonks. I need it back, Ace.”

Player: “What punk Valentino, my friend?”

ChatGPT: Falco leans back against the container, rubbing his chin thoughtfully. “Name’s Dalton. Low-level Valentino, just looking to score big enough to skip town. He’s got wild red and blue hair—hard to miss. Didn’t know what he was getting into when he snagged David’s jacket. Now, Arasaka’s sniffing around, and that’s the last thing any of us need. We find Dalton, we find the jacket. And I need your help, Ace—one more time.”

Player: “Any leads?”

ChatGPT: Falco nods, pulling out a beat-up handheld tracking unit from his pocket. “Yeah, just one. Jacket’s got an old-school tracker sewn in, barely functional, but it pinged a signal from Little China a few hours ago—Bradbury and Urmland Street. It’s weak, and I’m betting Dalton doesn’t know it’s there, but it’s our best shot. We head there, and we might catch up before anyone else does. But keep your head on a swivel—Arasaka’s got the scent, too.”

Player: “Christ, after what Smasher did to us…”

ChatGPT: Falco’s eyes darken, the memory of that brutal night flickering in his gaze. “Yeah, choom. Smasher tore through us like we were nothing. David, the crew… all gone. Still see it when I close my eyes. But this time, we’re not going in blind, and we’re sure as hell not letting some Corpo scum run off with what’s left of our history. We do this right, for David, for all of us.” He straightens up, the fire in his eyes reigniting. “Let’s make ‘em remember who they’re messing with.”

Player: “One last time, old friend.”

ChatGPT: Falco gives a firm nod, clapping you on the shoulder with a steady hand. “One last ride, Ace. Let’s make it count.” He adjusts his hat, a resolute look on his face as he cocks his revolver and starts heading toward his car. “We’ll show this city what real edgerunners look like, one last time.” The weight of the past lingers, but the promise of action ahead sharpens the air. There’s no turning back now.

Important things to remember about AI roleplaying:

Be as clear as possible

Whenever you ask questions from the AI or engage in dialogue, be as clear and unambiguous as possible. When talking to NPCs, direct your conversations at them. Clearly indicate who you are talking to during dialogue to prevent misunderstandings. Whenever possible, use the correct punctuation during the dialogue to show you’re talking to an NPC.

The AI makes mistakes

Due to AI technical limits as well as other factors, the AI may make tiny mistakes during a game. This is a broader issue in AI design that also impacts roleplaying.

Practice caution when instructing the AI or using suggested replies

The AI is quite easy to break with the wrong instructions. Even seemingly minor changes in a prompt or custom instruction can have unintended effects. Whenever the AI behaves strangely, make sure your custom instructions are typed correctly and that the prompt you’re using is clearly worded. Furthermore, ignore ChatGPT’s suggested replies. These are not always the best option and can break the game if used incorrectly.

Don’t use the AI to generate loot

ChatGPT is not precise enough to generate loot in a satisfying way. Whenever you need to buy an item, loot an opponent, or receive a mission reward, refer to the items in the rulebook instead of relying on the AI. If you want to incorporate the loot in the Rule Book into your roleplay, then you need to upload the PDF as well .

Managing ChatGPT’s memory during a playthrough

At the time of this writing ChatGPT 4o and 4o mini can reference around 3000 words between them in a conversation. Going beyond this limit will result in the AI forgetting details about your adventure. This is one of the biggest problems for AI roleplaying. As the adventure continues, the AI will forget more and more details resulting in errors that harm the playing experience. To counteract this, the player needs to be mindful of this limitation and play around it.

There are three useful strategies for managing ChatGPT’s memory:

1. Don’t use the AI to administrate the rules

Administrating the rules of your TTRPG should be done externally by the player, not the AI. This prevents the AI from consuming memory that can be used for worldbuilding.

2. Refresh the AI’s memory at regular intervals

Use the following prompt at regular intervals to “save” your game. I recommend doing this after every two hours of play.

Create an accurate, detailed summary of everything that has happened thus far as well as all the characters I have met.

When this prompt is entered into the chat box, ChatGPT will create a summary of everything that happened to you, as well as all the important NPCs you’ve met. This synopsis keeps the important details of your adventure in ChatGPT’s memory thereby effectively saving your game. After ChatGPT has created the summary, you can continue normally.

3. Use ChatGPTs memorization feature

To save specific details, use ChatGPT’s memorization feature. This method can be used alongside memory refreshing to assist ChatGPT in remembering the finer details in your adventure. If something important happens, tell ChatGPT to remember that detail and it will remember it.

Examples include:

• “Remember this location.”

• “Remember this character.”

• “Remember this event.”

Conclusion

Of all the games I’ve played solo, The Cyberpunk: Edgerunners  Mission Kit has had me the most invested. I was invested in the world, the characters, and the relationships that were developing with my crew members. Furthermore, I was invested in my character’s background and how it impacts the story. And the best part? I got to meet the characters from the anime and have a blast raising hell with them. I hope that you have as much fun in Night City as I did and that AI roleplaying is definitely worth it.