Universe Rules

How This Chapter Is Organized

The five core subsystems below are presented in dependency order: each one builds on the one before.

• Cover & Legends is the identity layer. It governs what happens when the fiction asks who does the world think you are, and what does it cost when that breaks?

• The Tradecraft Pool is the preparation layer. It governs what happens during the Intel Phase before the operation begins, and how the team’s homework converts into dice at the table.

• Assets & Networks is the human-source layer. It governs recruited contacts, pressure on them, and what happens when they are used.

• Compromise & Attribution is the consequence layer. It governs the slow accumulation of being noticed, how hostile services decide who did the operation, and how that accumulation ends careers.

• Era Framing is the tuning layer. It modifies each of the above, and a few core rules, differently depending on which era the table is running.

After the subsystems, a short section on worked examples walks through how they feel at the table, followed by homebrew guidance and the usual In This Chapter sidebar.

A Note on Precedence

Where Division 5 gives a Universe-specific procedure, limit, or option, use it for games set in Division 5. The Exodus NDS Core Rules remain the baseline everywhere this book is silent.

Inside this book, Chapter Eight defines the default Division 5 procedures. Talents, archetype signatures, equipment, adversary traits, mission playbooks, and era rules may modify those procedures within their specific scope. When a specific rule changes a default procedure, use the specific rule for that case. Otherwise, use Chapter Eight.

Core Rules Override Reference

This table collects the intentional changes Division 5 makes to the Exodus NDS baseline. If a rule is not listed here or explicitly called out elsewhere, use the Core Rules.

Subsystem 1: Cover & Legends

Every operative of Division 5 carries a stack of Legends: prepared cover identities, each with documents, a backstory, a supporting cast of fabricated contacts, and an exit protocol. A Legend is a container the operative pours themselves into for the duration of an operation, not a disguise.

Legends are the currency of the profession. An operative survives by building them, maintaining them, and burning them at the right moment; a Legend treated as permanent eventually costs the operative everything.

Building a Legend

Every operative begins play with three Legends: one Thin, one Standard, and one Deep. Each Legend is defined by:

• A name. The identity’s full name, including any middle names or patronymics the era demands.
• A cover profession. What the Legend does for a living. Tied to the era (a Belgian commercial agent in 1943, a Canadian trade attaché in 1974, a regional risk analyst in 2024).
• A location of residence. A specific flat, address, or city. The Legend is registered here.
• A paper trail. Documents, records, and, in Nightscreen, a digital residue appropriate to the depth.
• A supporting cast. A handful of fabricated contacts: a “sister in Geneva,” a “university classmate in Toronto,” a “former employer in São Paulo.” These contacts can be reached for verification.
• A depth tier. Thin, Standard, or Deep. This drives the mechanics.
• A Burn Track. Three, six, or nine boxes. This is how the Legend dies.

The starting stack is built during character creation with the Handler’s or Quartermaster’s help as narratively appropriate. Archetype signatures and talents may modify it (the Face begins with two Standard Legends instead of one; the Ghost’s Deep Legend begins with one extra box on its Burn Track).

The Burn Track

A Legend’s Burn Track is marked every time the Legend takes a hit: a discrete event in the fiction that erodes the cover. Common triggers:

• A hostile service runs a name check on the Legend and the check returns inconclusive or suspicious. (1 hit)
• The operative is photographed by opposition surveillance while using the Legend. (1 hit)
• The operative fails a Deception, Charm, or Tradecraft check while in the Legend in front of hostile scrutiny. (1 hit per unmitigated Failure, to a maximum of 2 per scene)
• A contact in the Legend’s supporting cast is interviewed by a hostile service and cannot be reached in time to coach. (1–2 hits, GM discretion)
• A biometric capture links the Legend to another identity. (Nightscreen: 2 hits minimum, up to full burn depending on the capture.)
• The operative uses the Legend in a context the Legend’s backstory cannot support. (“The Belgian commercial agent has no reason to be in Damascus this month.”) (1–3 hits, GM discretion)
• The operative burns the Legend deliberately for narrative advantage. (see below.)

When the Burn Track is fully checked, the Legend is burned: permanently compromised, cataloged by the opposing service, and unusable by any D5 officer, anywhere, ever again.

Legend Depth Tiers

Two archetype modifiers lengthen these tracks: the Face’s Cover Artist adds +1 to any Legend the Face uses (Standard 7, Deep 10), and Ghost Deep Legends run to 10. These variant lengths appear in the cooling and wear-band tables that follow.

Deliberate Burns

An operative may deliberately burn a Legend to claim a one-time narrative benefit. Mechanically, the operative may either:

• Add +2 Boon dice to any one critical check while in the Legend during the current scene, OR
• Escape a capture or compromise attempt that would otherwise succeed, by sacrificing the cover identity entirely (the hostile service believes it caught and identified the Legend, not the real operative).

A deliberate burn fills every remaining box on the Burn Track.

The Cooling Period

A Legend that has not been burned but has taken hits can sometimes be rested: set aside for a cooling period during which the opposition loses interest, surveillance rotates away, and the file drops down the stack. Cooling uses the Legend’s current Burn Track length after all modifiers: Standard 6, Deep 9, Face Standard 7, Face Deep 10, Ghost Deep 10, and so on.

Thin Legends do not normally cool. They are too shallow and too disposable; once a Thin Legend has attracted meaningful scrutiny, the proper procedure is to stop using it and build another one.

Cooling is a downtime activity and must be GM-approved. Cooling runs in the background; the operative is typically using other Legends during this period. A tired Legend with 1 residual hit can be used again, but it is no longer pristine: that residual hit remains until the GM agrees that the cover has been absent from hostile attention for years, or until a specific operation repairs the support structure.

Rebuilding a Legend

A new Legend can be built in a burned Legend’s place during downtime. The rebuild cost depends on depth, era, and the operative’s support structure.

Thin Legends rebuild in a week or less and require one GM-approved downtime scene. In Nightscreen, Thin Legends require an additional check: Cyber or Forgery (Average) to plant the necessary digital residue.

Standard Legends rebuild in one to three months of downtime. This is a major offscreen commitment. The operative typically hands over field duties to a teammate during the build, though the GM may run short scenes during the rebuild that advance other plot threads.

Deep Legends rebuild in years. An operative can ordinarily only begin building a new Deep Legend between campaigns, or as the setup for a major time-jump in the fiction. In Nightscreen, Deep Legends are built by the Garden, the D5 sub-unit whose entire function is long-lead identity construction, and the operative’s role in the build is to cooperate with the Garden’s instructions for the duration, not to drive the process.

Legend Use in Play

When the operative enters a scene in a Legend, they name the Legend at the table. Every social or tradecraft check in the scene that would invoke the Legend’s identity is made as the Legend: the operative’s real Characteristics and skills still apply, but the story is the Legend’s story.

Some talents and archetype abilities interact directly with Legends. The Handler’s Safe House talent sets up a local presence tied to the Legend currently in use. The Forgery skill can shore up a tired Legend for a single scene (Hard check; one hit cleared; does not work on burned Legends).

Carrying more than one Legend into a scene is possible but expensive. An operative can plausibly switch Legends once per operation, typically during a transition between phases, but cannot cycle through three identities in an afternoon without drawing attention. The GM should apply Bane dice to any check that relies on a Legend the operative has only been in for less than a scene.

Legend Qualities

Legends have optional qualities: small flavor tags that tilt a Legend toward particular strengths or vulnerabilities. A Legend can have up to two qualities. Qualities are chosen at build time and are semi-permanent (can be revised only when the Legend is rested).

Aged

The Legend is old. The paper trail and supporting cast predate the operative’s career. Effect: opposition background-check attempts against the Legend take 1 additional Bane die (harder to penetrate). Cost: the Legend cannot be refreshed with modern details without a full downtime rebuild.

Networked

The Legend has an active, reachable supporting cast. Phone numbers are answered. Emails get replies. Effect: contacts within the Legend may be reached for corroboration once per operation (treat as a successful Assist on a single check). Cost: maintaining the contacts is a downtime expense: reduce the next operation’s Requisition Budget by 1 for each session in which the Legend remains actively supported.

Transactional

The Legend has an audit-ready financial history: bank accounts, tax filings, credit records consistent with the stated profession. Effect: Economic or customs checks made against the Legend gain +1 Boon die . Cost: real money flows through the Legend; if it burns, a portion of the operative’s funding burns with it.

Clean

The Legend has no prior operational history. It has never been used by any D5 officer before. Effect: the Legend’s Burn Track is treated as +1 larger for the first operation. Cost: after that first operation, the quality is spent forever. If the temporary box is occupied when the quality expires, the hit remains and may immediately burn the Legend if the normal Burn Track is now exceeded.

Bruised

The Legend has history, it has been used before, perhaps by another officer who retired or was lost. The file is known to the opposition in outline. Effect: the Legend’s first hit each operation is free (a “known face, nothing new”). Cost: if the Legend takes three or more hits in a single operation, the old file is connected to the new activity and the burn cascades, mark 2 additional hits immediately.

Specialized

The Legend has a cover profession with unusually narrow expertise: a restoration conservator who works only with 17th-century Dutch oils, say. Effect: +1 Boon die to any professional check inside the specialty; the Legend’s presence is plausible in specialized venues. Cost: −1 Boon die to checks outside the specialty (the specialist is out of depth anywhere else).

Subsystem 2: The Tradecraft Pool

The single biggest tonal difference between Division 5 and a generic action engine is that this game rewards preparation. The Shadow War, the Long Dark, and Nightscreen share a common operational truth: the hours spent before the operation decide the outcome of the hours spent during it. The Tradecraft Pool is how that truth becomes a mechanic.

The Operation Structure

Every Division 5 adventure, a mission or operation, is structured into four phases. The Tradecraft Pool is built in Phase Two.

The Intel Phase is where the team earns the dice it will spend during Execution; a team that skips it starts Execution with an empty pool.

Building the Pool

During the Intel Phase, each player character makes one or more Intel checks against a GM-set difficulty (typically Average or Hard, with the occasional Severe for a hostile-source query). Intel checks are framed narratively:

• “I read the dossier on the target carefully.” → Knowledge (Geopolitics) or Perception.
• “I conduct a static surveillance on the meet location the night before.” → Surveillance.
• “I reach out to my network for a background read.” → Streetwise or Negotiation.
• “I set up and test our comms and escape tech.” → Cyber (Nightscreen) or Cryptography.
• “I verify the asset against the Ledger: has anyone worked him before?” → Knowledge (Geopolitics), with the archivist’s mandatory caveat.
• “I prepare the Legend’s paper trail for scrutiny.” → Forgery.

Each check that succeeds contributes to the Tradecraft Pool. Boons on the check convert generously: the Intel Phase is the phase where Boons are most valuable. The conversion table:

Pool Size Limits

A well-prepared team enters Execution with a Tradecraft Pool of dice. Every PC may contribute to it; the pool is shared across the team.

The pool has a maximum size. The cap is set by the number of PCs in the operation, to keep a large team from generating an uncappable resource.

Dice earned beyond the cap are lost. The Intel Phase ends when the team declares it ended, or when the GM signals that further Intel work triggers hostile notice (see Soured Intel, below).

Spending Pool Dice

During Execution, any PC on the operation may spend dice from the pool on any check they are about to make. The declaration happens before the dice are rolled. Spent dice are added to the roller’s pool for that check only.

Pool dice may also be spent for specific narrative declarations: explicit prep that turns out to be relevant:

• “I brought a second radio.” → Spend 1 Boon die from the pool to establish that a second comms unit is stashed in the operative’s gear. (GM may require it to be plausible for the Intel Phase’s framing.)
• “I already met the front-desk clerk last week.” → Spend 1 Ability die from the pool to establish that a minor NPC is friendly to the operative.
• “The guard shift changes in four minutes.” → Spend 1 Boon die from the pool to impose a useful timing beat on the scene.
• “There’s a back way out.” → Spend 1 Ability die from the pool to establish an exit route the scene didn’t previously have.

These are narrative spends. The GM is free to impose reasonable limits on scope: a pool die cannot establish a tank in the next room, cannot make an asset trust the PC who has never met them, cannot rewrite the target’s location.

The Pool Clears at End of Operation

Unused pool dice vanish at the end of Exfiltration; each operation’s pool is rebuilt from scratch during its own Intel Phase. The pool represents current, operation-specific intelligence, and it ages out with the operation that produced it.

Soured Intel

The Intel Phase is not risk-free. Researching a target is an act of intelligence, and hostile services watch for exactly such activity. If the Intel Phase drags, or if Intel checks produce Calamity results, the opposition becomes aware that someone is asking questions. The GM has two levers.

The Soft Signal. If Intel Phase checks accumulate two or more Calamities across the phase (from any PCs, in any order), the opposition is alerted in general terms. Apply +1 Bane die to the first check of Execution: the operation begins with the target slightly more vigilant than planned.

The Hard Signal. If Intel Phase checks accumulate four or more total Calamities, or a single Calamity on a critical check (GM discretion), the opposition has acquired specifically targeted awareness. The target is forewarned. One of the following applies, GM choice:

• The target has moved. Execution opens with the target in a different location than briefed.
• The meet is a trap. The asset is under opposition control. The first meet of Execution reveals itself as a setup (the pool still spends; the scene still plays; the trap is what the scene is).
• Surveillance is live. The opposition has mounted active surveillance on the team’s approach. All Tradecraft and Surveillance checks during Execution add +1 Crisis die .
• The file is on the desk. A dossier on the team (or, more specifically, on the Legend they are using) is active with a rival service. Compromise accumulates faster for the remainder of the operation.

Types of Intel Checks

As a GM reference, a typology of the most common Intel check framings. Each corresponds to a skill, and each contributes to the pool in slightly different flavor.

Dossier Dive

Skill: Knowledge (Geopolitics) or Perception. The character reads the file. Historical context, personal background, known associates, prior operations. A thorough Dossier Dive can add or remove GM-discretion beats in Execution based on Boons and Banes.

Static Surveillance

Skill: Surveillance. The character watches the target or the location from cover. Timings, patterns, personnel, security routines. Builds location-specific pool dice: these can be spent only on checks that take place at the observed location.

Active Surveillance

Skill: Surveillance or Tradecraft. The character follows the target or a person of interest. Riskier than static: Calamities here more often result in a Soft or Hard Signal. Produces more Boon dice when it succeeds.

The Quiet Ask

Skill: Streetwise, Negotiation, or Charm. The character queries an asset, a former colleague, or a professional acquaintance. Produces Boon dice and occasionally a narrative detail. Overused, the Quiet Ask is the fastest route to a Soft Signal.

The Bench Check

Skill: Cyber, Cryptography, or Forgery. The character tests gear and prepares materials: comms, encryption, documents, forged credentials. Produces Ability dice reliably. Does not risk Soft Signal unless the check involves an outward-facing query.

The Ledger Query

Skill: Knowledge (Geopolitics). The character consults D5’s institutional memory. Has anyone worked this target before? What happened? The Ledger Query produces unusually high-quality dice but is limited: the institution allows one Ledger Query per operation, by policy, and a second query in the same operation requires director-level approval (GM call).

OSINT Sweep

Skill: Knowledge (Geopolitics), Cyber, or Perception. The character searches open sources: public records, corporate filings, leaked databases, press, social media, procurement notices, academic citations, satellite imagery, and commercial data brokers. In Nightscreen this is a primary Intel Phase activity. Produces broad context and technical signatures. A Calamity means the search pattern itself was noticed by a platform, contractor, or hostile analyst.

The Walk-Through

Skill: Tradecraft. The character physically rehearses the operation’s routes, timings, and dead drops. Produces Boon dice specifically for Execution scenes that follow the rehearsed path. If Execution deviates, the Walk-Through’s dice are less useful (GM discretion; halve the dice contributed).

Subsystem 3: Assets & Networks

An Asset is a person outside Division 5 who can provide access, information, cover, logistics, or misdirection. Assets are not equipment. They have fear, loyalties, blind spots, pressure points, and lives the operation can ruin.

Every ongoing asset has a small profile:

Recruiting an Asset

Recruitment requires a scene. The player describes the approach and rolls the appropriate skill:

• Charm for affection, admiration, friendship, or seduction.
• Negotiation for money, exchange, leverage, or transactional recruitment.
• Leadership for ideology, cause, duty, or institutional appeal.
• Streetwise for criminal, social, or local-network recruitment.
• Tradecraft for cold approaches, dead-drop recruitment, or pressure-sensitive clandestine contact.

The GM sets difficulty by risk and resistance.

On a success, create the asset at Trust 1, Exposure 0, Capability 2, and Access 1 unless the fiction says otherwise. Spend Boons to raise Trust or Access by 1, reduce starting Exposure by 1, or define a useful detail about the asset’s environment. A Breakthrough may start the asset at Trust 2 or Access 2.

On a failure, the asset does not commit. Boons may still create a future opening. Banes add Exposure, create a rumor, or reveal the approach to a third party. A Calamity means the target reports the approach, is already controlled, or becomes a hostile dangling asset.

Long-Term Cultivation

Some assets are not recruited in one scene. Long Dark campaigns especially should make room for cultivation arcs that stretch across several operations: the professor who must be brought from sympathy to action, the border guard whose first favor is small, the hostile officer who becomes a mole only after years of testing.

Use the Cultivation Loop when the relationship is important enough to play across operations.

Each stage normally requires one meaningful scene or downtime action. A fast recruitment can compress stages; a patient mole run can stretch one stage across a season.

Trust Drift

At the end of any operation involving an ongoing asset, check whether Trust or Exposure should drift.

Trust cannot rise above 5. Exposure cannot fall below 0. If Trust ever falls below Exposure at the end of an operation, the GM may test for panic, defection, double handling, or confession.

Running an Asset

Once per operation, each ongoing asset may be tasked for one specific job. The Handler describes the task and either rolls a relevant PC skill or makes an Asset Check using the asset’s Capability as the Ability dice rating. Add Boon dice equal to Trust if the task aligns with the asset’s motive. Add Bane dice equal to Exposure if the task risks discovery.

Common asset tasks:

On success, the asset provides the benefit. Additional Boons may add a Tradecraft Pool die, reduce the next attribution check by one hostile Boon, or increase Trust by 1 if the handler protects the asset. On failure, the benefit is partial, delayed, or false. Banes increase Exposure by 1. A Calamity increases Exposure by 2 and creates an immediate scene: interrogation, panic, disappearance, hostile surveillance, or betrayal.

Trust and Exposure

Exposure rises when the asset is tasked, watched, interrogated, used in a public incident, connected to D5 gear, or forced outside their routine. Exposure clears only through downtime, relocation, a successful counter-surveillance scene, or a deliberate false trail. Most assets clear 1 Exposure after a full quiet season. Nightscreen assets often clear none unless their digital records are actively scrubbed.

Burning an Asset

At Exposure 5, the asset is burned. Choose the outcome that fits the fiction:

• Captured. The hostile service has them. The handler marks +1 Compromise unless the asset plausibly knows nothing linking back to the handler.
• Doubled. The asset remains in place but now serves another master. The GM may use them as a hostile asset until discovered.
• Dead. The asset is killed, disappears, or is made an example. The handler marks +1 Strain immediately and may mark Compromise if the death points back to the team.
• Fled. The asset survives but is unusable until relocated. A rescue or exfiltration operation may convert the burned asset into a recurring ally.
• Public. The asset’s exposure becomes a scandal, trial, leak, or broadcast story. Run Attribution immediately.

An asset can be deliberately burned for a major benefit: automatic access to one scene, a clean false trail, a decisive warning, or a life-saving betrayal of the opposition. Deliberate burn is never free. The handler marks +1 Compromise or loses 1 Trust with every other asset in the same network, GM’s choice.

Era Asset Treatment

Sample Asset Profiles

Asset Advancement and Retirement

An asset who survives repeated use should change. Do not let an important asset remain a static stat line for a whole campaign season.

At the end of an operation in which the asset mattered, choose one if the fiction supports it:

• Improve Access. The asset reaches a better file, post, network, route, office, platform, or person.
• Improve Capability. The asset learns the work, becomes braver, or receives training and tools.
• Increase Trust. The handler protected them, told them the truth, paid them fairly, or honored a promise.
• Reduce Exposure. The team cleaned the trace, moved the meet, built a false trail, or gave the asset quiet time.
• Change Pressure. The motive shifts: money becomes ideology, ideology becomes fear, fear becomes loyalty, loyalty becomes resentment.

Capability or Access should not rise more than once per season unless the asset is becoming a major supporting character. Trust can move faster; Exposure often moves fastest of all.

Retire an asset when their story has reached a stable end: resettlement, asylum, witness protection, a normal life under another name, a promotion that makes tasking impossible, prison, death, or double handling. A retired asset can return as a favor, a vulnerability, a senior NPC, a hostile witness, or the reason a new operation exists.

Subsystem 4: Compromise & Attribution

The Compromise Track records long-term hostile awareness (for how it divides labor with Strain, see Compromise vs. Strain, below). Every PC has one. It runs from 0 to 5. At 5, the character is burned (see Burned, below).

This is the primary long-term pressure in a D5 campaign: the world’s growing awareness of a specific person who should not be known at all.

The Levels

How Compromise Accumulates

Compromise goes up when specific events happen in the fiction. The GM marks the track when:

• A Legend is fully burned in a way that links to the real operative. (+1) The opposition has a file with a real face on it now.
• The operative is photographed, captured on CCTV, or biometrically scanned in a context hostile services will review. (+1 per distinct capture; Nightscreen only.)
• The operative is identified by a hostile asset under interrogation. (+1 or +2, GM discretion.)
• A mole inside D5 passes the operative’s file to an outside service. (+1 at minimum; +2 if the mole is handled by a peer service rather than a rival.)
• The operative fails a critical Stealth, Tradecraft, or Deception check with a Calamity. (+1)
• A personal connection, family, former spouse, long-term friend outside the service, is interviewed. (+1)
• The operative’s service record is exposed publicly. (+2; rare but campaign-defining.)
• The GM offers a Fortune Point for exposure. The GM may offer a player 1 Fortune Point to accept a Compromise mark when the fiction supports it: a witness gets a better look, a file links two appearances, a family connection is tugged, or a quiet mistake becomes part of the hostile record. If the player accepts, mark +1 Compromise and add the Fortune Point to the player side of the Story Point pool. If the exposure is campaign-defining, the GM may offer +2 Compromise with the player’s explicit agreement. This is not a Peril payout; Peril belongs to the GM side of the pool.

Conversely, the operative may accidentally accumulate Compromise from cumulative small mistakes, even without a single triggering event. The GM may mark the track at narrative break points to represent erosion: typically between sessions, and typically no more than +1 per session of inactivity on the cleanup front.

Passive Compromise Pressure

Passive Compromise is the background clock of hostile territory. It only applies while the operative is under active hostile exposure: operating in a watched city, using a Legend in front of institutions that can check it, staying in a theater after a public incident, or remaining connected to a live hostile investigation. It does not punish quiet months in a stable cover with no active pressure.

The GM may adjust the rate by the Threat Environment modifiers later in this chapter. Passive pressure should be announced at Tasking or Intel, not sprung as a surprise during Debrief.

How Compromise Clears

Compromise does not clear easily. It is the slow pressure of the campaign, and clearing it is itself a campaign activity. Common clearing mechanisms:

• Long Downtime (three months or more, no operations, out of theater). −1 Compromise. Once per downtime. Requires the operative to be genuinely off the board: no side work, no field checks. Most campaigns permit this once per “season.”
• A New Deep Legend. −1 Compromise when a new Deep Legend is brought online. The operative “becomes someone else.” Costs the standard Deep Legend build time.
• Permanent Relocation. −2 Compromise when the operative’s entire operational theater changes. This is a major campaign beat: effectively retiring a character’s current life and starting a new one. Most campaigns permit this once.
• A Contra-Op. −1 Compromise when the team runs an operation specifically targeting the hostile file itself: corrupting the dossier, compromising the analyst who maintains it, or leveraging a source inside the rival service.
• An Asset’s Sacrifice. −1 Compromise when an asset takes responsibility for operational activity the operative was actually responsible for. The cost falls on the asset.
• A Cleaner’s Favor. −1 Compromise when the Cleaners’ Guild removes the surveillance infrastructure around the operative. This costs money and creates a debt with the Cleaners’ Guild, which is not trivial to repay.

Burned

When the Compromise Track hits 5, the operative is burned. They are pulled from the field immediately: before the end of the current operation if the burn is triggered mid-op, at the next operational break if triggered between. Depending on the fiction, this means one of the following:

• Disavowed. The operative is cut from the service’s rolls. They go grey (see Chapter Two). Surviving but no longer on the rolls, they may return as an NPC, a consultant, or a ghost character in a later campaign.
• Captured. A hostile service has taken them. Subsequent campaigns may include an operation to recover them; some such operations succeed.
• Killed. The burn includes a death, in the field, in custody, or in transit.
• Reassigned. The operative is pulled to a non-field role (Handler, Analyst at a desk, instructor). They survive, they know too much to be let go, and they never go into the cold again.
• Retired to Grey. The voluntary out. The operative leaves the service intact, accepts the loss of the cover life they had, and disappears into a small life somewhere the file does not reach.

Compromise and the Team

Compromise is individual, not team-wide. Each PC tracks their own number.

However, Compromise has contagion effects. A PC at Compromise 3 or higher brings operational risk to the rest of the team:

• Compromise 3+: Teammates working alongside this operative in theater suffer +1 Bane die on Tradecraft and Surveillance checks that involve the compromised operative’s active presence.
• Compromise 4+: Teammates working alongside this operative in theater risk Compromise transfer: a GM-discretion event in which the team’s collective surveillance footprint causes a teammate to mark +1 Compromise of their own at the end of the operation.
• Compromise 5 (burned): If the burn is public (captured, identified, named in the press), every teammate in the operation marks +1 Compromise of their own. A D5 burn is a service-wide event.

This is why high-Compromise operatives are sometimes benched by the team: sent out of theater, held in reserve, kept away from active operations. It is also why the team has a pragmatic incentive to help a senior operative work their way toward retirement rather than let them go out hot.

Compromise vs. Strain

The core rules include a Strain track that abstracts short-term stress and wear. Strain still exists in Division 5: an interrogation, a firefight, an extended chase can all generate Strain. Compromise is separate. Strain clears within a session or an encounter. Compromise persists across sessions, across operations, across Legends.

As a general rule of thumb:

• Events that can be shaken off by morning are Strain.
• Events that leave a record somewhere in the world are Compromise.
• Events that could conceivably do both (a prolonged interrogation that yields no break but produces surveillance photographs, say) may generate both, at GM discretion.

Strain heals. Compromise does not heal without work.

Core Wound and Strain recovery never clears Compromise. A character can be physically rested, medically stable, and emotionally functional while still sitting at Compromise 4 because hostile services have not forgotten them.

Compromise Events: A Short Gallery

To help GMs calibrate, a set of common events and their Compromise impacts:

Attribution Procedure

Attribution is the post-operation question: who does the world think did this? It is resolved during Exfiltration or debrief, after the team has left the scene but before rewards, fallout, and downtime.

Run Attribution whenever an operation leaves enough residue for another service to analyze: public violence, abandoned gear, dead bodies, a vehicle chase, digital traces, a burned Legend, an exposed asset, official records, or a hostile witness. Clean operations may skip the roll; messy operations should always run it. The procedure exists to make gear choices, vehicle choices, asset treatment, and cleanup matter after the bullets stop.

Step 1: List the Signatures

The GM lists every meaningful signature the operation produced.

Step 2: Build the Hostile Attribution Pool

For each signature, add dice to the hostile attribution pool.

Pool cap: the hostile attribution pool normally caps at 6 Ability dice , 4 Boon dice , and 2 Mastery dice . Capstone failures, public scandals, and season finales may exceed this at GM discretion.

Step 3: Add Obscuring Dice

The team adds opposing dice for cleanup, misdirection, and deniability. Use the best relevant PC contribution, not every PC contribution.

If the team did no cleanup, add no defensive dice; a fast exit on its own obscures nothing.

Step 4: Roll and Read

The GM rolls the hostile pool against the defensive dice. Read the final outcome:

If the team planted a specific false trail, a Calamity result from a defensive Crisis die means the hostile analysis catastrophically follows that false trail. This is not a new core result type; it is the D5 attribution meaning of a defensive Crisis die generating Calamity during this procedure.

Attribution never has to reveal the PCs’ real names to matter. “A British-adjacent off-book service with access to SOE legacy kit” is a dangerous attribution even if no one knows the operative’s birth certificate.

Era Modifiers

Subsystem 5: Era Framing

Each era modifies the core rules in specific ways, tuned to make the genre feel right at the table. The modifications are small but load-bearing: a table that runs Shadow War with Nightscreen rules will feel wrong, even if no single player can articulate why. Run the tunings as written for a first campaign. After ten sessions, a table can start adjusting them, dialing Compromise faster or slower, raising or lowering the Shadow War damage modifier, or opening Cyber partially in Long Dark for a technothriller campaign; none of those adjustments are wrong, but none should be made before the table knows how the default feels.

Era I: The Shadow War (1939–1945)

The Shadow War is deadlier than the other eras. Combat has teeth, cover is cheap, and capture is usually terminal.

• Firearm damage is +1. Firearms in this era are reliable killers. Increase the listed damage of Ranged and Sniping attacks by 1 before applying Resistance.
• Capture scenes resolve in one scene, not a chain. Once a Shadow War operative is captured, the GM does not run a week of interrogation at the table. The scene is: arrive at the facility, endure, escape or break. One session, maximum. Operatives who break give up one Legend, their operation, and one piece of sensitive intelligence per failed Resilience check.
• The Cyber skill is unavailable. Tables may repurpose Cyber as Signals Engineering with GM approval, but the default is that the skill does not apply.
• Legends build twice as fast. Thin Legends in an afternoon, Standard Legends in weeks, Deep Legends in months. The tradecraft of fabricated identity is less sophisticated in this era: and the verification capabilities of opposing services are less sophisticated.
• Legends burn twice as fast. The other side of the coin. Each hit on a Legend’s Burn Track checks 2 boxes instead of 1. A Thin Legend is 3 boxes total, which under Shadow War rules survives one mistake. For cooling periods and condition bands, count checked boxes, not the number of triggering events.
• The L-pill exists. Every operative carries one unless narratively justified otherwise. Its use is a table-level beat; the GM adjudicates as the moment requires.
• No drones. No digital or satellite SIGINT. No facial recognition. The technology layer is analog. Radio intercept, traffic analysis, cryptography, and direction finding absolutely exist; what does not exist is the automated global collection layer of Nightscreen. This removes several pressure types from the game: and replaces them with others, mostly involving informers, captured radios, and direction-finding vans.

Taken together, these tunings make a Shadow War campaign short, sharp, and high-attrition. Characters cycle. Players build new ones. The Compromise Track matters less than the Legend Burn Track, and an operative’s career is measured in operations rather than years.

Era II: The Long Dark (1946–1989)

Tradecraft is king. The Long Dark is the era the system was most explicitly built for, and its modifications reinforce that bias.

• Intel Phase upgrades preparation. Each Intel check that succeeds upgrades one die in its Tradecraft Pool contribution: a contributed Boon die becomes an Ability die , and a contributed Ability die becomes a Mastery die . If every contributed die from that check is already upgraded as far as it can go, add +1 Boon die instead. A check with a Breakthrough contributes its Mastery die as normal and then upgrades one other contributed die if possible.
• Moscow Rules. Every PC and every named NPC with intelligence-service background is assumed to operate under the doctrine; a scene that violates Moscow Rules is a scene the operative is choosing to violate them in. The mechanical reroll, however, belongs to the Moscow Rules talent in Chapter Five, which modifies this baseline: operatives without the talent still know the doctrine, they simply do not get the talent’s extra die control.
• Combat is lethal and politically expensive. Firearm damage is standard, not elevated. Violence in the Long Dark is professionally dangerous because every service studies deviations from routine. The GM should impose +1 Compromise for a firearm discharged in a civilian zone, regardless of outcome, and should ask what official, witness, hostile service, or asset now has to be handled.
• Legends build at standard rates. Thin in a week, Standard in months, Deep in years. Long Dark is the baseline against which the other two eras flex.
• Legends cool reliably. The cooling-period table is honored. A Long Dark campaign can afford a deep Legend rotation, burn one, rest another, build a third, across a long campaign.
• The Walk-In is a meaningful narrative beat. Tables should prepare at least one Walk-In event per campaign season. A civilian or low-level opposition officer approaches the team offering defection. The vetting is a mini-arc. Some Walk-Ins are real. Some are provocations. The table learns to tell the difference.

Run with these tunings, a Long Dark campaign is patient, procedural, and morally dense. Operatives have long careers. The Compromise Track matters more than the Legend Burn Track, and the game is about sustained relationships: with assets, with handlers, with a city. Violence is available, but it echoes. Information is the goal state.

Era III: Nightscreen (1990–Present)

SIGINT dominates. The tools are commensurate with the threats: a well-equipped Nightscreen operative carries more capability in a briefcase than an entire Shadow War cell possessed.

Use the technology bands from Chapter Two. Early Nightscreen (1990-2001) uses records, pagers, faxes, airline systems, CCTV islands, and early mobile logs more than biometrics. Networked Nightscreen (2001-2012) uses counterterror databases, mobile metadata, and fragmented online platforms. Saturated Nightscreen (2013-present) uses the full phone, cloud, biometric, and platform-correlation pressure described below.

• Compromise accumulates faster. Biometric captures, facial recognition hits, and digital residue events each trigger a mark under How Compromise Accumulates. A modern Western capital’s baseline CCTV coverage adds +1 Bane die to any scene requiring the operative to be genuinely unwatched. At Compromise 3+, the operative’s real face is in opposing services’ facial-recognition databases; any public scene in a covered city imposes a Tradecraft check (Hard) to avoid a Compromise mark.

• The Cyber skill is primary. Intrusion, encryption, OPSEC, and counter-intrusion are as important as Tradecraft for most operations. Every team should have at least one PC with Cyber 2 or better. A team without it has to compensate: usually by leaning heavily on the Quartermaster or hiring out to contractors, both of which introduce risk.

• Drones are an environmental hazard. Overhead surveillance is no longer exotic. A modern operation in most theaters assumes the possibility of drone coverage. PCs may spend Tradecraft Pool dice to scout for drones; the GM may introduce drones with Peril Points to make an otherwise-safe scene urgent.

• Legends build harder but last longer when they are built well. A Deep Legend in Nightscreen is a major investment: years of work, serious institutional resources. In exchange, a Nightscreen Deep Legend can survive events that would have burned a Long Dark Deep Legend. The Burn Track is standard (9 boxes); the quality of the Legend is higher per box.

• The Garden handles Deep Legend construction (see Rebuilding a Legend), and the Garden sets the pace. GMs should run at least one scene per campaign that takes place inside a Garden consultation, it is some of the best texture in the era.

• The attribution war shapes mission goals. Most Nightscreen operations have an attribution objective in addition to the operational objective. A mission that completes its primary goal but leaves a traceable attribution trail is a partial success at best. The table should adjudicate attribution at end-of-operation: Boons earned during Execution can be spent to obscure attribution; Banes on critical checks leave it exposed.

What emerges is a campaign that is technologically dense, geopolitically messy, and morally ambiguous. The enemies are harder to draw. Private actors matter. The traditional nation-service axis is still present but no longer solely operative. Of the three eras, this is where the Compromise Track bites hardest.

Era Framing Summary

A one-page reference for the GM to pin near the dice tray.

Cross-Era Campaigns

Some tables will want to run campaigns that span multiple eras: inheriting files from one to the next, playing flashback operations, running a cold-case investigation that dips into 1978 to make sense of 2024. These campaigns work, with care.

The default guidance:

• Pick one era as primary. The table’s default pacing, default gear, default threat profile lives there.
• Use flashback or archive scenes for the off-era work. A flashback is a one-session or part-of-session beat, not a sustained secondary campaign.
• Translate Legends across eras carefully. A Long Dark Legend is not a Nightscreen Legend. The paper trail must be re-established. Era-specific qualities (Transactional is easier in Long Dark, Aged is native to Shadow War) shift meaning.
• The Compromise Track resets per era. A character who was Compromised 3 in their Long Dark campaign does not inherit Compromise into a Nightscreen cold-case frame unless the fiction requires it. Most cold-case fictions do require it: the whole point is that the past isn’t past.

Cross-era work is the advanced mode for Division 5. A first campaign should pick one era and stay in it. The rhyme structure in Chapter Two (1943 dead drop = 2024 encrypted post) means the skills carry forward when the table is ready.

Worked Examples

Eight worked examples follow. Each example compresses a typical scene into a paragraph of fiction and a walk-through of the mechanics. Read these before the first session to get the feel of how the procedures behave at the table.

Example 1: Cover & Legends: The Interrogation Room

Vienna, 1974. Analyst-Handler Josef Seitz, D5 cover identity “Karl Brandt, Swiss trade representative.” Under questioning by a Stasi HVA officer who has recognized something wrong about the documents.

Seitz has been using the Karl Brandt Standard Legend for eighteen months. Before this meet, the Legend had 6 boxes on its Burn Track and 1 hit already (a prior surveillance photograph that had taken the Track to 1/6). The HVA officer asks Seitz to account for a three-month gap in his travel records from 1971.

The GM calls for a Deception check at Hard difficulty. Seitz rolls. He gets 1 Success, 1 Boon, 2 Failures, and 1 Bane. The Success and Failures net to 1 Failure; the Boon and Bane cancel. Net result: 1 Failure. The check fails.

Mechanical consequence: The Legend takes 1 hit (Burn Track now 2/6). Because the Boon canceled with the Bane, Seitz gains no side benefit from the roll. He fails cleanly: the HVA officer does not prove the truth, but the gap remains suspicious and the Legend is weaker.

Narrative beat: The HVA officer files a partial report on Brandt. The Legend is not burned, but it is exposed. Seitz lives to meet the officer again next month, under heightened caution.

If Seitz had burned the Legend deliberately: Seitz could have announced, before the check, that he was burning the Karl Brandt Legend to escape capture. The Burn Track would fill immediately; Brandt would be cataloged as a dead file by the HVA. Seitz himself would slip away in the confusion, his real identity still intact. Karl Brandt would never be used again, anywhere.

If the check had gone worse: 2 Failures, 1 Bane, and a Calamity would have triggered: the Legend takes 3 hits; the Calamity is narrated as the HVA officer linking Karl Brandt to a photograph from an unrelated operation, which takes the Burn Track to 4/6 and puts Seitz at +1 Compromise (a real face is now in the HVA file). The scene ends with Seitz needing extraction.

Example 2: The Tradecraft Pool: The Rooftop Chase

Dubai, 2024. Four-PC team. Primary objective: intercept a courier exchanging a dataset on a rooftop in Business Bay.

Intel Phase. The team has 90 minutes of table time. They make five Intel checks:

1. Analyst runs a Dossier Dive on the courier. Success + 2 Boons. Contribution: +1 Ability, +2 Boon.
2. Wetworker runs Static Surveillance of the building. Success. Contribution: +1 Ability. (Location-flagged.)
3. Face runs a Quiet Ask with a former colleague at a private-intelligence firm. Success + Boon. Contribution: +1 Ability, +1 Boon.
4. Quartermaster runs a Bench Check on gear. Success + Breakthrough. Contribution: +1 Mastery.
5. Analyst runs a Ledger Query. Failure, no Boon. Contribution: Nothing. GM narrates: “The Ledger is silent on this courier. Either he’s too new to be in it, or he’s too well-hidden.”

Tradecraft Pool at start of Execution: 3 Ability, 3 Boon, 1 Mastery. (Team cap is 10 dice for 4 PCs: well within limits.)

Execution scene. The courier makes the exchange. The team moves to intercept. The Quartermaster declares a narrative spend: “The comms jammer I built: it’s pre-positioned on the roof. One Boon die from the pool to establish it.” The pool now reads 3 Ability, 2 Boon, 1 Mastery.

The Wetworker makes a Ranged check to disable the courier’s escape route (destroy an access door lock). Base pool: Agility 3 + Ranged 2, built by the Core Rules as 2 Mastery and 1 Ability. Wetworker declares: “I’m spending the location-flagged Ability die : I prepared for this.” Pool adds 1 Ability for the check. Wetworker rolls and scores 3 Successes, 1 Boon. Net: Success, +1 Boon. The door jams. The courier is cornered.

Later in the scene, the Analyst declares: “There’s a back way out of this roof complex: I found it on the floor plans.” One Ability die from the pool establishes the back exit. Pool now reads 1 Ability, 2 Boon, 1 Mastery. The Analyst uses the back exit to reposition for a clear shot with a tranquilizer pistol.

Exfiltration. The team takes the courier out clean. Unused pool dice, 1 Ability, 2 Boon, 1 Mastery, vanish at the end of the operation. The team earned them; they did not need them all. Next operation starts fresh.

Example 3: The Compromise Track: The Border Crossing

Modern era. PC: Handler Ana Dvořák, Compromise 2 (“Surveilled”).

Ana is attempting to cross from Serbia into Hungary under her Standard Legend (“Margita Horvat, Croatian film location scout”). The border is biometrically-equipped. Standard passport control.

Base state: Compromise 2, so +1 Bane die on Tradecraft and Social checks in theater. Standard Legend with 2 hits (4/6 remaining).

The check: GM calls for a Tradecraft check (Average) to pass scrutiny. Ana’s pool: Savvy 3 + Tradecraft 2, built by the Core Rules as 2 Mastery and 1 Ability. Cover adds 1 Boon (Networked quality on the Legend allows a phone call to a “client” that corroborates the story). Compromise adds 1 Bane. Final pool: 2 Mastery, 1 Ability, 1 Boon, 1 Bane.

The roll: 2 Successes, 1 Boon, 1 Failure, 1 Bane. Net: 1 Success, no extra Boon/Bane. The Legend passes scrutiny.

Consequence: The crossing completes. But the biometric kiosk does its job: the capture is logged. GM rules that because Ana is at Compromise 2, and because the Legend is tied to the real operative (Networked quality makes the Legend more easily linked in some cases), the capture advances the track.

Mark +1 Compromise. Ana is now at Compromise 3 (“Flagged”).

Downstream effect: For the rest of the campaign, Ana has +2 Bane dice on all active-cover checks. Her handler pulls her out of Eastern European rotation for six months. She uses the downtime for a Long Downtime Compromise clear (−1, back to 2). Her Margita Horvat Legend remains at 2 hits and can return after six months of rest, but using it again in the same theater before then would carry obvious risk.

The GM notes in prep that, at Compromise 3, Ana’s real face is in at least one hostile service’s facial-recognition system. The next time she appears in a covered city, a Tradecraft check (Hard) will be triggered.

Example 4: Era Framing: The Same Operation in Three Eras

The operation: a long-running asset has been compromised. The team must extract them across a border before a hostile service can sweep them up.

Shadow War (Berlin, 1944). The asset is a Wehrmacht officer with knowledge of a weapons program. The extraction route is a Lysander pickup in a field outside Wolfen. The team has three nights to do it. Combat-heavy: the Wetworker is expected to engage. Legends burn fast: the team will probably be on Thin covers, which break under one mistake. Capture would be terminal. The operation ends in either the field or the interrogation room; there is no middle outcome.

Mechanical shape: Small pool, fast burn, high damage, short arc. The session runs in 3–4 hours. Attrition is real.

Long Dark (East Berlin, 1974). The asset is a Stasi officer who has been working for D5 for a decade. The extraction is through a prepared route under the Wall: a tunnel, a third-party diplomatic vehicle, a forgery-heavy document package. The team has two weeks. The Intel Phase runs deep. The Face spends three sessions cultivating a border guard’s sister. The Handler verifies the asset is not under HVA control (they are not, but it takes doing). Combat is the contingency everyone hopes to avoid; a firefight at the border turns the operation into survival, cleanup, and inter-service damage control.

Mechanical shape: Deep pool, slow burn, careful tradecraft, long arc. The operation plays out over three or four sessions. Compromise accumulates from the pattern of life being established during Intel.

Nightscreen (Minsk, 2024). The asset is a junior GRU officer selling to the West. The extraction route is an SUV-and-chartered-aircraft combination, but the attribution layer is the real problem, the team must extract the asset in a way that leaves a different service’s fingerprints on the exfil. Biometric kiosks at the airport. Drone surveillance above the route. A Garden-built Deep Legend is burned on the operation; it has to be, because the Legend needs to survive the biometric capture. The Compromise Track accumulates heavily, for every PC in the operation, +1 to +2 depending on how clean the exfil is.

Mechanical shape: Mid-pool, high Compromise cost, technically dense. Session runs 5+ hours. The team spends as much Pool on obscuring attribution as on the physical extraction.

Same operation. Three entirely different feels. This is the era framing working as designed.

Example 5: Soured Intel: The Too-Clean Courier

Long Dark. The team is preparing to intercept a courier in Prague. The first three Intel checks all ask the same question from different angles: where does the courier sleep?

The Analyst succeeds on a Dossier Dive. The Handler fails a Quiet Ask with a Calamity. The Face succeeds with a Boon at a hotel bar. The Wetworker then runs Active Surveillance and also rolls a Calamity.

Two Calamities in the Intel Phase trigger a Soft Signal. The GM does not cancel the operation. The courier still appears. But Execution begins with +1 Bane die on the first approach check because the courier has noticed interest around his hotel.

The players later learn why: the Handler and Wetworker both touched the same hotel clerk from different covers within four hours. The courier is not brilliant. The team’s prep was simply too narrow. If the team had produced a fourth Calamity or a Calamity on a Ledger Query, the GM could have escalated to a Hard Signal: the courier moves rooms, changes the meet, or arrives with a surveillance team.

Example 6: Attribution With Dice: The Hotel Killing

Nightscreen. The team kills a hostile private-intelligence contractor in a hotel service corridor under Lethal Force Pre-Authorized conditions. The body is left to look like a criminal robbery.

The GM lists signatures: hotel CCTV, a suppressed pistol casing recovered from a drain, a missing phone, one witness who saw a cleaner’s cart, and a false wallet planted on the body.

Hostile pool:

• CCTV is Biometric or Digital Match: +1 Ability; Nightscreen would add +1 Mastery if tied to a real identity, but the team wore viable Legends.
• The casing is Contextual: +1 Ability.
• The witness is a Soft Signal: +1 Boon.
• The planted wallet is a false trail and does not add hostile dice.

Hostile pool: 2 Ability, 1 Boon.

The team adds defensive dice: successful Tradecraft cleanup removes the hostile Boon; Forgery on the planted wallet adds +1 Trial; one Tradecraft Pool Boon is spent during Exfiltration and converts to +1 Bane against the hostile roll. Final opposed pools: hostile 2 Ability against defensive 1 Trial, 1 Bane.

The hostile roll succeeds with no extra Boons. Result: probable attribution. Because the Legends stayed intact, the GM marks one hit on the exposed Legend rather than +1 Compromise to a real operative. The public story is a botched robbery. The hostile-service story is better: an off-book team killed the contractor, but they do not yet know which one.

Example 7: Mole Hunt: The Canary Route

Long Dark. Three station officers had access to an extraction route. The route appeared in KGB hands within a week.

The GM gives the suspect map finite boundaries:

• Officer A has unexplained money and was near the archive late.
• Officer B has a romantic contact attached to a Soviet cultural office.
• Officer C altered the access log but has the cleanest personnel file.

The players review the file with Knowledge, interview support staff with Charm, and run a controlled leak through Tradecraft: three false route variants, each given to one suspect. A week later, KGB surveillance appears on the variant given to Officer C.

The reveal is an earned confirmation, not “C is guilty” by fiat. The altered access log was not a coverup of the leak; it was C hiding that they had copied the file for their handler. Officer A’s money was from a black-market camera sale. Officer B’s romance is real and dangerous but not the mole channel.

Mechanically, the controlled leak is a Tradecraft check at Hard difficulty. Success identifies the mole route. Boons let the team preserve the station’s public integrity. Banes mean C knows the test was a test and moves to burn an asset before the team can act.

Example 8: Shot Advantage: The Balcony Window

Long Dark. Vienna, 1978. The team has authority for lethal force only if a hostile courier is positively identified leaving a diplomatic reception with a stolen microfilm canister. The Wetworker, Mara, is in an apartment hide across the street with an M21. The Face and Analyst are inside the reception. The Handler is in a car two blocks away.

Authority and condition. The brief authorizes the shot only if the courier is confirmed carrying the canister and crosses the balcony alone. Killing the wrong person would be unauthorized wet work and a major diplomatic scandal.

Building the shot. Mara has Sniping 3, so her Shot Advantage cap is 3.

• The Analyst identified the balcony routine during Intel with Surveillance: +1 Shot Advantage.
• The Quartermaster prepared the hide and sightline: +1 Shot Advantage.
• The Face confirms the courier’s lapel pin and the canister handoff inside the reception with Perception: +1 Shot Advantage.

Mara is now at the cap: Shot Advantage 3. More preparation would still matter, but it cannot raise the pool beyond 3.

Pressure. The courier steps onto the balcony, but a waiter follows two seconds behind. Mara can fire now with the waiter near the line of sight, wait one Aim maneuver, or abort. She waits. The GM does not add Shot Advantage because she is already capped, but advances scene pressure: the waiter asks the courier a question, and the balcony window is closing.

The shot. Mara spends 1 Shot Advantage for +1 Boon die and holds 2 Shot Advantage. She rolls Sniping against Ranged Defense and succeeds with a Boon. Because the target is unaware, the position is prepared, the weapon is appropriate, and Mara has 2 held Shot Advantage, the Clean Kill rule applies. The courier drops inside the balcony shadow before the room understands what happened.

Aftermath. The team still has to survive the operation. The shot creates a ballistic signature, a dead diplomat-adjacent courier, and a reception full of witnesses who heard glass break. Exfiltration runs Attribution. Because Mara waited for the waiter to move, the GM does not add Public Heat from a visible bystander death. Because the weapon was military and the scene was diplomatic, the hostile pool still receives an Institutional signature.

Alternate outcome. If Mara had fired immediately and failed with Bane, the courier would live, the waiter would be wounded or able to identify the shot direction, and the Alert Clock would jump to Exposure. If she had aborted, the operation would fail cleanly: no kill, no public scandal, and the next operation would become a courier recovery or mole-hunt problem.

A Final Word on Bending the Engine

The five core procedures in this chapter are the reason Division 5 is a Division 5 game. Each carries a load none of the others can pick up: the lie at the center of the profession, the mechanical reward for patience, the human cost of intelligence, the price of a long career, and the distinct texture of each era.

A table that wants a lighter version of the game can dial any of the five down: smaller pool caps, slower Compromise accumulation, simpler Legend tracking, softer asset burns, or looser Attribution. No table should remove one outright. The rest of the book assumes all five are running.