Lesson 1 locked your fantasy. Lesson 2 gave your project folders and naming rules. Now you prove the mission works in space: greybox is where stealth either feels fair or falls apart, long before polished art arrives.

This lesson is about readability—cover density, lanes, risky crossings, and traversal beats that match your written success conditions. You are not lighting a trailer yet. You are building a spatial script players can read at a glance.

Lesson objective

By the end of this lesson you will have:

1. A greybox level saved under /Game/Maps that walks the primary beats from your Lesson 1 outline
2. Documented metrics (door width, cover height, risky corridor length) so Lesson 4 movement tuning has targets
3. A first-pass navigation volume mindset (even if full AI lands in Lessons 5–7) so blockout does not fight patrol design later

Step 1: Open the right map and set scale truth

1. Open L_Mission01_Setup from Lesson 2 or create L_Mission01_Greybox under /Game/Maps/M01_Blockout.
2. Drop a scale reference: mannequin, 140cm capsule sketch, or a known prop cube at real-world units (Unreal centimeters).

Rule: if doorways feel wrong, fix scale before you duplicate fifty corridors. Players forgive grey materials; they do not forgive unclear player size.

Step 2: Choose BSP, meshes, or hybrid blockout

Pick one primary method for the slice so merges and diffs stay predictable. Mixing all three without rules creates invisible “one-off” walls that break AI bounds later.

Step 3: Place beats before decorations

Walk your Lesson 1 beat list in order:

• Entry – where tension starts.
• First decision – risky vs safe path.
• Objective A – interaction volume placeholder.
• Complication – patrol cross or camera sightline puzzle.
• Climax – detection pressure peak.
• Exit – win trigger placeholder.

Use notes (placeholder actors, World Outliner folders, or a parallel Design_Notes text asset) so level designers syncing tomorrow know which BSP chunk is story-critical.

Step 4: Cover rhythm and sightlines

Stealth greybox is rhythm:

• Full cover (player fully hidden from a standing guard at key viewpoints).
• Soft cover (breaks silhouette but not guaranteed safety).
• Exposed beats where the player must commit to movement.

Spacing checklist:

• Place two reference guards as static meshes at head height to judge sight cones informally. You will replace them with AI in Lesson 5.
• Avoid long symmetric corridors with identical clutter; players cannot remember risk. Break symmetry with one readable landmark per leg.
• Keep crouch-route options plausible if Lesson 1 promised them—mark ceiling height and vent volumes now.

Pro tip: if a crossing feels “free,” shrink cover or lengthen exposure. If it feels hopeless, add a second lane or a timed distraction volume placeholder (Lesson 8 will formalize gadgets).

Step 5: Traversal cadence (not parkour soup)

Define how often the player:

• Stops to observe (perch spots with sightlines to patrol loops).
• Commits to a sprint or slide (length and hazard marked).
• Re-routes when detection raises stakes (alternate loop connects back to objective).

If traversal is flat, stealth becomes waiting simulator. If every room demands tricks, you blew the Lesson 1 session budget. Aim for peaks and valleys on a timeline you can narrate in under two minutes verbally.

Step 6: Navigation and collision sanity pass

You do not need polished behavior trees yet, but you do need:

• Walkable floors with continuous collision (no hairline cracks that snag CharacterMovement).
• Door frames wide enough for capsule + strafe.
• Volumes for areas you already know will be off-limits or trigger zones (name them Trigger_DetectionTutorial, Volume_Extraction, etc.).

When the nav mesh preview returns (“AI” class features later), you want the walkable surface to match the player graph you just playtested on foot.

Step 7: Lighting stub (greybox only)

Add a neutral skylight + directional combo so silhouettes read. Avoid final color grades.

Goals:

• Players see enemy heads against backgrounds.
• Cover edges cast readable shadows, not black mud.

If you cannot read silhouettes now, Lesson 12 lighting will not save you.

Mini challenge

Record a two-minute flythrough (or third-person run) of the greybox hitting every Lesson 1 beat in order.

• No combat required.
• Stop the recording if you get lost—that is a layout bug, not a minimap problem.

Troubleshooting

Summary

• Greybox is mission proof, not art polish.
• Cover, traversal, and beats must align with Lesson 1 before AI polish.
• Scale and navigation choices now prevent expensive fixes in Lessons 6–7.

Next, Lesson 4: Player Stealth Controller Baseline wires crouch, noise, and visibility rules so traversal promises turn into mechanics.

Further reading

• Unreal Engine documentation hub – level design and geometry tools reference.
• Unreal Engine guides – companion material on this site.
• Unreal Engine 5.5 packaging troubleshooting – keep handy for final shipping lessons.

FAQ

Should I greybox the whole hub world?
No. One mission slice that matches your session length target is enough until vertical slice review.

When do I replace static guard meshes?
After navigation and sightlines feel fair on foot. Lesson 5 introduces perception; use placeholders until then.

Do I need Landscape for this course slice?
Only if your Lesson 1 fantasy demands it. Many indoor stealth slices use static floors first.

When your greybox tells the story without UI, you are ready to script movement that respects the space. Bookmark this lesson and move to Lesson 4 once your two-minute run respects every beat in your design doc.