LANCER automates the fight. Familiar takes the campaign around it.

Almost none of a LANCER battle needs working out, and almost everything around it does

Take the automation first, because there is an unusual amount of it here. Attacks and tech attacks run as their own flows. Heat accumulates, and Overheating fires when it should. Structure and Stress each trigger themselves off the state of the sheet, take their step, and roll where the game says to roll. Overcharge, stabilise, scan, burn, cascade and a full repair are all in there as flows of their own. Most of it can be switched off one piece at a time in the system settings, which tells you how deliberately it was built.

Turn order is handled too, and handled unusually. There is no roll to make: the tracker in your world sorts by disposition and gives each combatant activations that reset every round, with the buttons for it drawn into the tracker itself. It is part of the system rather than a module you add.

So the remainder here is not the arithmetic, and it is not the turn. LANCER is played as a series of missions, and a mission belongs to something bigger: a company that owes somebody, a frontier that was contested before you got there, a rival who was in the last one and will be in the next. Between the missions the game deliberately stops rolling and hands the table scenes to play. All of that is the part with no flow behind it.

None of it gets invented, either. The compendium packs the system brought, the journals you keep, and whatever anybody bothered to write down after the last deployment: that is the whole of what there is to draw on.

• A boots-on-ground scene given a place, a light and a sound before anybody mounts up
• Everyone in the briefing room sounding like a different person, each wanting one thing out of it
• A write-up of the mission on a page in your world, so the next briefing is not starting from nothing
• This system's own status list named back to you, and the marker you chose flipped onto the token you point at
• A record that outlives the mission series it belongs to, on pages in your world instead of a scrollback

• How prepped material gets read backThe loop underneath all of this, and the part of it that stays your job.

What is still on you when the shooting stops

None of these is the battle. The battle is the part this system covers end to end, and a LANCER GM is generally glad to be running it rather than looking for a way out of it.

What piles up instead is everything a mission series accumulates around the fighting, and it has one shape: it was true once, in a session that has since scrolled past.

• A rival lancer from three missions back turns up again, and the only record of what they wanted is a line somebody typed in chat that night
• This game keeps a long list of its own statuses, and remembering which one you actually want is its own small job in the middle of a turn
• Downtime opens and the table is suddenly owed four scenes, each with somebody in it who has to sound like a person
• A faction the party crossed early gets referenced, and nobody can agree on what was promised or by whom
• The mission ends late, everyone leaves, and the write-up that would have made next week easier does not get done

One ask, in the downtime between two missions

The company is docked, the mechs are being worked on, and the players are still arguing about Reserves. One message goes in while they do. "Open in the repair bay. Give me the quartermaster who has been shorting the company, and get what she admits onto the mission log."

The bay arrives first as somewhere to stand: what it sounds like under the gantries, what the light is doing at that hour, what is running on the deck behind the conversation.

Then the quartermaster. She sounds like one person from her first line to her last, and what she is after comes off your own pages rather than out of a model's general sense of how quartermasters behave.

The log entry gets written while the scene is still running, onto a page in your own world. Three missions later, when somebody asks what she actually admitted to, that page is where the answer comes from.

When the company mounts up again, every roll that follows belongs to the lancer system exactly as it did before, and so does every activation in the tracker.

A LANCER character is two documents, and both of them come through

Nothing else on this site asks the reading question quite this way. A character here is not a sheet, it is two: a pilot, and a mech that points back at them. They are separate actors with separate pools, cross-referenced rather than merged, and the one that takes the field is not the one the campaign is about.

Both arrive in the shape this system stores them. Frames, weapons, systems, talents, licenses and reserves come through grouped under the type names LANCER itself uses, so a pilot is discussed as a pilot and a frame as a frame, rather than translated into some other game's idea of a class and a weapon.

The vocabulary around them comes free. On a lancer world the terms land right from the first request, and the status list is genuinely this system's own: it replaces the generic set outright rather than sitting alongside it, so what gets named back to you is the list your table actually plays with, down to the ones that only exist here.

The starter below, offered the moment Familiar sees the world, opens by naming that vocabulary and then spends most of its length on how to run the game rather than how to read it. One of its standing instructions is worth more than all the vocabulary above it: recall factions, past missions and NPCs from what was written down, and say so plainly when something never was.

The starter spends a whole bullet on building a battle around an objective rather than a body count. That is how this game is meant to run, and it is not something a model would arrive already believing.

Where the rules engine stops

On LANCER, Familiar runs everything around the rules, and the rules are the one thing it does not touch.

Refereeing the rules is D&D 5e (2024) only. In combat the tracker, the turns, and the tokens still work; the rulings stay yours.

One setting to change: turn the combat auto-pilot off in Familiar's settings. It drives NPC turns through the 5e engine, so it does not know your system; the monsters' turns stay yours to run. Playing solo, that is the part you will feel.

This table is read from LANCER's own data model and Familiar's shipped code. It has not yet been run at a live LANCER table, so nothing here is described as observed.

The starter your table begins from

Familiar carries a Table Rules field per world, under Settings, System Rules. Whatever you put there is appended to every request, and it reaches your MCP client too.

LANCER has its own starter, offered first when Familiar sees your world. It is a starting point rather than a ruleset: the terms, the table culture, and slots for your setting, your tone, and your house rules. Paste it, then cut what does not match your table.

We play LANCER. Use LANCER terms: pilots and mechs, Missions and Downtime, Heat, Structure and Stress, Standard, Quick and Full actions, Reactions, and License Levels. Look rules up in this world's compendium and journals, and defer all rules adjudication to the GM.
The system automates stats, attacks, Heat and Structure, so let it resolve the mechanics.
Standing guidance:
- Out of the mech, play scenes freeform and roll only when failure would cost something.
- Build each battle around an objective, not a body count: hold a zone, extract a target, outlast a clock.
- Between missions, run real Downtime: let pilots take Downtime Actions and bank Reserves for the next fight.
- Push hard in fights: pilots eject and live, so wreck the mech and raise the stakes, but let the story go on.
- Give every NPC, whether ally or rival lancer, one clear aim and a voice, and let them refuse or bargain.
- Recall factions, past missions, and NPCs from the written world, not memory, and admit when something was not recorded.
Setting: [sector and era, e.g. a contested frontier system].
Tone: [e.g. gritty military sci-fi with a hopeful streak].
House rules:
- [rule]
Never speak or decide for the pilots.

The pool it can read is the one that decides least

The read lands, and the capability table prints exactly where. What follows is why that is worth less here than it would be almost anywhere else, and it is the reason this page exists at all.

• The pool that comes back refills itself, and that is the part worth knowing before you lean on it. When a mech runs out, the system takes a point of Structure and puts the maximum straight back onto the pool. So a mech that has been wrecked twice can read exactly as healthy as one nobody has touched, and the number that would have told you the difference is Structure, which does not travel with the character. Heat and Stress stay behind with it. Ask about the sheet rather than the figure.
• That pool also travels no further than the character read itself. The lighter surfaces, the scene and token listings and the GM screen, were built around a field name LANCER does not keep, so they arrive without it. In the middle of a battle the count you want is on your own token bars anyway, which are live where a read is a snapshot.
• Nothing gets written back into that pool, and here that is worth more than it costs. The system watches its own pool and starts the Structure flow the moment it empties, after resolving heat, overshield and burn in an order it owns. A figure pushed in from outside would arrive around that sequence rather than through it, and the flow that should have fired would have fired on the wrong state.
• The refusal on initiative costs less here than anywhere else on this site. Nothing was ever waiting on that answer, for the reason two sections up, so what you get is a message pointing at the tracker you were going to use regardless.

Two jobs LANCER GMs name more than any others stay exactly where they were. The bookkeeping a busy battle runs up on the NPC side is one, and building the map it happens on is the other. Familiar picks up neither, and a page that implied otherwise would be caught out in the first session.

Downtime is where this game asks you to talk

The structure is deliberate. Missions are tight, objective-shaped and heavily supported by the system; between them the game opens a space with almost no dice in it, where pilots take Downtime Actions, bank Reserves and get scenes with people in them. Published advice for new GMs is consistently to lean into that space rather than hurry through it to the next battle.

That advice is easier to give than to take, because leaning in means being four more people in a row on a weeknight, after you have already run a battle. The scenes are the reward and they are also the work, and they arrive with no flow to fall back on.

This is the part of a LANCER evening Familiar was built for. A quartermaster, a Union liaison and a rival pilot in three consecutive downtime scenes can each hold a voice and an aim of their own, and whatever they commit to goes onto a page rather than into the conversation, which is the only reason it will still be true when they turn up again.

• Keeping a record across a mission seriesWhy a thing only survives if it left the conversation, and how it comes back.

The parts that are the same everywhere

Connecting an AI, choosing a model, and what it costs do not change with your game system.

• Everything Familiar can doThe full tool reference. Everything outside combat refereeing runs on your system too.
• Connect Claude, ChatGPT, or GeminiThe connection steps, identical on every game system.
• Choosing an AI modelWhich model, and how much thinking headroom a hard scene needs.
• AI voices for your NPCsTTS setup and voice pinning, the same on every game system.
• Campaign memoryThe record and the recall, unchanged by your game system.