Role-playing Games Stack Exchange is a question and answer site for gamemasters and players of tabletop, paper-and-pencil role-playing games. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

When I design adventures, I often come up with situations where the NPCs could realistically thwart the PCs quickly by sounding an alarm. This is especially common in science fiction settings, where the NPCs have access to surveillance cameras, radars, motion detectors, and other advanced technology. This makes stealthy missions difficult to pull off, because the PCs can be revealed in a matter of seconds if they make even a single mistake.

How do I keep NPCs from just sounding the alarm right way, without suspending disbelief too much?

share|improve this question
1) Watch the PCs storm into a secure and hostile area without bothering to prepare for security and hostility. 2) Let the NPCs sound the alarm. 3) Watch the PCs die in a hail of blaster fire/captured and sent to the prison planet/fed to a pet bagorth. 4) Ask them to roll up new characters who will be less suicidal. 5) Profit. (Said somewhat tongue-in-cheek. ;-) – SevenSidedDie May 2 '12 at 18:09

15 Answers 15

up vote 45 down vote accepted

The thing with science fiction settings is that when technology creates a problem for the PCs, it can also generally create a solution. Part of the heist genre is figuring out what the alarm systems are and getting your hands on the gadgets you need to neutralize them. Other valid tactics include social engineering tactics like bribing or impersonating the camera-watchers. The trick is getting players who realize that if they go up against alarm systems, they can lose very suddenly, and that therefore they need to do their homework and plan things out rather than rushing in blind.

Alternatively, if your PCs are working for an organization or higher power and don't have the means or wherewithall to deal with electronic security (no hacker or engineer-types), it's possible that "dealing with the alarm systems" is handled by someone or something else. If you go this route, you can also use it to build tension; "We can get you a four minute window of inactivity on the perimeter defenses. Don't waste it." Then you run into a complication on the way through and you're in jeopardy from the security systems, but it's not immediate / game-over jeopardy.

share|improve this answer


In most contemporary or near future settings, any time there is any sort of alert, there will also be an incident report. If a guard has to fill out a report every time he hits the alarm, s/he will make darn sure s/he has a situation requiring all that extra work.

Your automatic alarm is going off? grab a hand-held and go check it out. Is someone crawling over the fence, or is that overgrown tree swaying in the breeze enough to connect to the electric fence? having to fill out ANOTHER form for that dumb tree is a waste of time (in the guard's mind), so he likely won't hit the alarm.

Also, tailor the security to what is being secured. The company paying for all of that security won't employ the latest in razor-wire, automated mini-guns of DOOM if all that is being stored is tomorrow's lunch at the cafeteria. Some GMs I know will tailor the security to the party's ability to infiltrate a facility, not what is being secured.

share|improve this answer
This is a fun idea and it seems especially suitable for games inspired by action movies, where the guards always seem to check things out by themselves instead of sounding the big alarm right away. – Jakob May 2 '12 at 18:13
Roll 1d6. 1: they freak out and sound the alarm without properly checking; 2-5: they investigate; 6: they can't be bothered ("That damned faulty sensor in sector 12 going off again… You want another coffee?") – SevenSidedDie May 2 '12 at 18:14
I love this because it's such a mundane rationale, but it really does make sense. It also jives with the notion that a complex system will throw up errors from time to time, and a security guard probably has dealt with such errors a few times and will therefore doublecheck before hitting the alarm. – Erik Schmidt May 2 '12 at 18:56

"Step away from that alarm, rookie. We don't need the 5-0 kicking in the door and making everybody play nice. When you've been here as long as I have, you'll realize that a night like this is a night to have some fun. Why did I sign up to be a security guard if not to lay down some pain every once in a while? Let's go kick some ass!"

share|improve this answer
Suddenly, I am eagerly awaiting the chance to surprise my players with what appears to be an infiltration scenario that is actually a "being stalked through dark, unfamiliar halls by a psycho killer with a blue uniform and too many quiet nights to study blueprints and set up lethal traps" scenario. I might not actually do that, but it has appeal… – SevenSidedDie May 2 '12 at 19:00

The very simple answer is: you don't.

Roleplaying is about making decisions that have consequences. As a Game Master, it's up to you to make those consequences interesting and fun, but that doesn't mean they should always be beneficial to the party or the characters that incur them.

If the players set off an alarm, then they will learn pretty quickly that they need to think smarter and prepare more for future encounters. As long as you give them options and allow for reasonable escape opportunities, ultimately the fate of the characters should be in their hands, not yours.

share|improve this answer

I think the single largest thing you can do to make sure that everything works smoothly is to make sure that everyone is on the same page as far as expectations. You are welcome to have any level of paranoia/security you want, as long as your players understand that going in. If they know it's there, they'll plan for it. If they're unaware for carefully crafted story reasons, it adds to the drama. If they just didn't know, that makes for a huge let down (most of the time).

I played in a sci-fi game where everyone on a planet wore Heart Monitors. One of the other players chose to be a killer robot. He wasn't aware that the moment he killed someone, he was causing an investigation. It escalated to a point where he should have just died to a police manhunt. All this was for smuggling a package that, if caught, would have resulted in a small reprimand. We had no idea of the setting, though. This happened in our first session. We never had a second session.

But, in the retelling, I want to illustrate that the expectations were what ruined that. The robot went in expecting to need to silence any observer. Turns out he made more of a mess by doing so. If he had known about the Heart Monitors, he would have changed tactics. And I think that is the extreme example of an alarm, but even that kind of immediate response could have been handled, with appropriate warning. Spoof the signal (even temporarily), or kill in such a way that the heart keeps going for a short bit, or something. Just make sure that the players know what's possible in the setting, let alone what they're facing at the current moment. If they know what's available, and the likelihood they might come across that tech, it's then their job to be prepared.

share|improve this answer
Welcome to the site! Excellent answer situated in personal subjective experience. – Brian Ballsun-Stanton May 2 '12 at 20:39
Great answer! It's important for the GM to be honest about what's going on. No player likes to hear "What, you didn't realize that he had the magical X that totally screws you over? The offhand comment the fisherman's wife said in the first session should have been enough of a clue! Too bad, roll new characters..." – Daenyth May 24 '12 at 14:20

For all but the most paranoid of enemies, the alarm can be a last resort. If the guards outnumber (or hell, are even just evenly matched), they may just decide to handle the problem themselves rather than press the big red button. I mean, what security guard isn't confident in their ability to get rid of people in one fashion or another? If they weren't then they wouldn't be guards!

The other aspect of this is that sounding the alarm not only tells other people that there's a problem, but that they need to drop what they're doing and solve it. Tripping the alarm grinds everything to a halt; you only need to set it off once when you don't need to have the full attention of your bosses, and Bad Guy Bosses (tm) tend to be, well, unpleasant to have the attention of.

share|improve this answer
This falls into "play the opposition stupid", which is sometimes good advice. After all, the GM is omnipotent and omniscient but the opposition isn't, so it's sometimes a useful brute-force emulation of limited information. (And how often have we lamented the players doing something that seems "stupid" from our omniscient GM view, that is really just because they're making the best choice with limited information?) – SevenSidedDie May 2 '12 at 18:50

There are a number of approaches you can take, depending on the specific campaign.

1. Remember that many places don't have an alarm to hit. A lot of interesting targets will not have an alarm to hit, or will only have a very basic one that requires some human to trigger it manually. Many places will not have one due to cost and complacency. Others, especially in a cloak-and-dagger type game, will not have one precisely because they have something to keep secret and will not have or do anything that can betray what is going on inside or attract outside attention. And of course if it is something along the lines of a "panic button", then all the PCs have to do is either avoid detection by humans or take the person down before they can hit the button. That type of alarm will make the stakes of being detected higher, but not change any fundamentals.

2. Preparation As lorimer mentioned, dealing with the alarm can be part of the game. If the PCs know to expect an alarm they can find some gadget (in sci-fi) or spell or amulet (in fantasty) that lets them get past it. Or they can find some way to disable it ahead of time. Or they can use social engineering to avoid the issues (of course the gaurds will politely turn off the alarm for the nice repairman, people can be bribed to disable them, etc).

3. Does the alarm actually matter? Sometimes an alarm can go off and it just doesn't matter, besides adding flavor. If, say, the alarm just calls out to normal law enforcement the response time could be measured in several minutes (possibly more if there is a big event tying up law enforcement's attention or this is a dystopian future where law enforcement is stretched past the breaking point...) The PCs might be able to ignore the alarm, do their thing, and get out before the cavalry arrives, the alarm just adds some urgency.

Even better, if the PCs have enough information, they might be able to use the alarms to their advantage. If the PCs have one team deliberately trip an alarm somewhere, it just might pull security away from their actual target while they go in. Even if they unintentionally trip a second alarm, they might have some time and face a reduced security response because of the first alarm.

share|improve this answer
Not to mention the PCs triggering the alarm, then scurrying to a safe spot. Keep doing that and all but the most hyper-paranoid will simply ignore the alarm after a few goes. – Vatine May 23 '12 at 12:12
to #3: recent sci-fi game's in-book fiction: "This isn't the ghetto, we only have four minutes tops from the time you draw a weapon until [local law enforcement] arrives. Move fast." MAKE the alarm not matter by being gone before it matters. – gatherer818 Jul 22 '14 at 21:58

If you're simply running a kick-in-the-door campaign where you don't want your PCs to worry about security systems, you have every right to simply ignore them. The problem that this causes, then, is that if you do decide to make the security system active later on in the plot, your PCs may balk because it's something you've implicitly removed from your world.

If you would like to maintain realism as well as the possibility for future security systems, you'll need some way to explain away security systems you don't want the PC to deal with. Some ideas:

  • The guild/corporation/military that the PCs are working for have assured them that they have taken out the target's alarm/security system. (Simple but requires that the PCs don't work alone.)
  • The PCs have done some research and found that the alarm/security system, even when active, has a flaw in its design. This could also be useful for railroading their approach of the target.
  • The PCs have bribed/blackmailed/recruited an inside man who will overlook their entry/disable the alarm/etc.

All of these reasons require a little bit of work or prior roleplay, but allow for both realistic PC actions as well as missions with unknown security systems and requiring extra caution, if you want to go for that.

share|improve this answer

Seven-sided was not all that wrong in his jest.

Your description sounds like you want to keep the realism and the tension and the feeling of reward. SO I recommend a few things.

  • Be clear before this phase of the adventure starts about what type of surveillance is used commonly based on the particular knowledge of the PCs. Thief types and tech types will know a lot, but others might as well based on what security set-ups have been used in areas they have frequented. Some prior knowledge gives them the chance to avoid a lot of this.
  • Have said surveillance/alarm work as it is supposed to around the PCs in other situations. Banks, guilds, ports, borders, shops...there are a ton of places you can show these defences working as they should before it becomes life threatening.
  • Make sure the existence of countermeasures, technological or magic or mundane, has been made available when needed. Players love having stuff. And stuff that helps them do stuff is normally stuff they want and that tells them they might need it. Or at least where to go back to when they need it.
  • And don't fudge it when it happens. You've done what you can do give them knowledge and tools. Up to them at this point. Smarter Players often come from deader characters.
share|improve this answer

I think it depends on what kind of game you want to play. If you are playing a gritty game where getting past security systems is a big part of the challenge during game play (like in Shadowrun for example) then I think you should pay close attention to the details and sound the alarms when the PCs make a misstep. However, if bypassing security systems isn't as important to the story you are trying to enjoy with your players then don't worry about the fine details. There is little reason to focus on being realistic when it doesn't support the story.

share|improve this answer

For a sci-fi setting, which was presupposed in the original question, make sure that one of the PCs is a Tech/Engineer/GreaseMonkeyType. She or He should be able with a very routine skill roll, to determine exactly what kind of surveillance system is being dealt with, and then with another skill roll of minor, moderate, or major difficulty, depending upon the sophistication of the systems in place, either to (1) disable the system entirely (the easiest solution), (2) fool the system into thinking no one is there so no alert is sent, but leave the system operating normally (this is a more difficult choice), or (3) send a false signal (this would be the most difficult of these three possible solutions). I'd allow the PCs to choose which method of disabling/subterfuge was being performed, and adjust the difficulty of the task accordingly.

This is equally do-able for a fantasy setting: If the alarm is a bell or gong that is located within sight of the PCs, a simple Silence spell will usually do the trick. If the room is being monitored via claivoyance/crystal ball/etc. then the PCs will require some advance notice of this monitoring. Judicious use of an Invisibility or other illusion spell, to either make themselves totally undetectable or else appear as creatures who belong in the area, should accomplish the goal. Teleporting past the room with the alarm would also work well, if that option is available to the PCs.

share|improve this answer

There are two flaws to every security system: the rigidity of electronic systems and the fallibility of human beings. Just because there are motion detectors and cameras doesn't mean the guards at the other end won't think its a false alarm if everything they are seeing looks like it. Likewise even in the even of a real alarm a well played bluff can easily get past the human element (attractive women faking drunk when a guard finds her when she shouldn't be in the embassy when there's a party going on). The best security systems have both but both can be systematically dismantled given enough time, thought, and money.

share|improve this answer

Usually in the games I'm a part of, the PCs' status as heroes sort of takes care of the alarm issue. If intruders easily slaughtered the 5 guys guarding your entrance, are you more likely to rush in there, or hunker down and take the best defensive position you can?

For more stealth or skill-oriented infiltrations, my approach is simply to allow multiple failures instead of one bad roll meaning a total failure. For example, a stealth failure could be described as, "A guard starts to turn a corner and his flashlight will shine directly on you. Everyone roll perception to see if they hear the guard in time to warn you." Chances are someone will make it, but then have some longer-term consequence. "You duck out of sight in time, but the guard thought he heard something suspicious and he's starting to check windows and doors for signs of forced entry."

share|improve this answer

In general my PCs know that any sneaking into fortress, castle, hi-rise, brothel is not going to be a cakewalk. I encourage them to approach it in a realistic manner such as scouting, social engineering, getting low level jobs to scout the inside, use of hidden mikes, tracking devices and robots.

Also for the most part I allow a little leeway such as if the PCs are sneaking in and forget to announce they are on the lookout for guards, detection devices and such. I assume they are since they are sneaking in.

If they are pretty much taking no pre-cautions feel free to bring the full effects of their actions to their attention. They may escape and find themselves having to break into the same place that now has triple the guards and double the detection devices.


The party had to break into a mansion to steal a prototype sniper rifle. The mansion was well guarded with troops, automated turrets, cameras, dogs, motion detectors, x-rays. Part of the party staked it out while the rest tried to find weakness in the defenses. After a few days they discovered a weak link. Any food going in was not X-Ray'ed that with the fact they knew the owner of the mansion was having a birthday soon. So they split up some got in via the birthday cake, another got in by kidnapping and replacing one of the drummers for the band playing, the rest got in as wait staff (they found out wait staff badges where low level but the easiest to reproduce).

share|improve this answer

In a cyberpunk scenario, you would typically have the physical infiltrators (call them grunts) working in tandem with cyberintruders who work the target's computer network.

The typical pattern works that the two groups are in communication (which might be interrupted during the scenario), the grunts would lay the groundwork by providing ports of access for the cyber intruders (e.g., spotting a communications cable and tapping into it, which might later be used to intercept an alarm message), and later the cyberintruders provide information to the grunts and deal with any "noise" they might produce.

The "alarm" is then a problem for the cyber PCs to deal with: stop the message from reaching HQ, ensure that HQ receives a reassuring flow of fake information, etc.

There is a lot of role-playing and problem-solving potential in this, but you have to ensure all the players have the same idea of what's going on. The couple of cyberpunk raid scenarios I've played fell to bits over this.

Funny, I don't think I've read a cyberpunk raid scenario in 15 years. They used to be all the rage.

share|improve this answer
Sounds straight outta Cyberpunk! Boy do I wish there was a 2020 game I could join. I bought an old used copy off amazon a few years back and it blew my mind. Most enjoyable rules book I've ever read if albeit less streamlined than I'd like. – Joshua Aslan Smith May 23 '12 at 15:40

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.