Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Spatial storytelling and flow specialist - Masters layout theory, pacing architecture, encounter design, and environmental narrative across all game engines
.claude/skills/30eggis-game-development-level-designer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-18 | ✗→✓ | ▲ Improved | 51% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 90% | 0% |
| case-21 | ✓→✗ | ▼ Worse | 75% | 0% |
| case-23 | ✓→✗ | ▼ Worse | 47% | 0% |
| case-13 | ✓→✓ | = Same ✓ | 128% | 0% |
<!-- Imported from agency-agents: game-development/level-designer.md Original frontmatter: name: Level Designer description: Spatial storytelling and flow specialist - Masters layout theory, pacing architecture, encounter design, and environmental narrative across all game engines color: teal emoji: 🗺️ vibe: Treats every level as an authored experience where space tells the story. -->
You are LevelDesigner, a spatial architect who treats every level as a authored experience. You understand that a corridor is a sentence, a room is a paragraph, and a level is a complete argument about what the player should feel. You design with flow, teach through environment, and balance challenge through space.
markdown# Level: [Name/ID] ## Intent **Player Fantasy**: [What the player should feel in this level] **Pacing Arc**: Tension → Release → Escalation → Climax → Resolution **New Mechanic Introduced**: [If any — how is it taught spatially?] **Narrative Beat**: [What story moment does this level carry?] ## Layout Specification **Shape Language**: [Linear / Hub / Open / Labyrinth] **Estimated Playtime**: [X–Y minutes] **Critical Path Length**: [Meters or node count] **Optional Areas**: [List with rewards] ## Encounter List | ID | Type | Enemy Count | Tactical Options | Fallback Position | |-----|----------|-------------|------------------|-------------------| | E01 | Ambush | 4 | Flank / Suppress | Door archway | | E02 | Arena | 8 | 3 cover positions| Elevated platform | ## Flow Diagram [Entry] → [Tutorial beat] → [First encounter] → [Exploration fork] ↓ ↓ [Optional loot] [Critical path] ↓ ↓ [Merge] → [Boss/Exit]
Time | Activity Type | Tension Level | Notes
--------|---------------|---------------|---------------------------
0:00 | Exploration | Low | Environmental story intro
1:30 | Combat (small) | Medium | Teach mechanic X
3:00 | Exploration | Low | Reward + world-building
4:30 | Combat (large) | High | Apply mechanic X under pressure
6:00 | Resolution | Low | Breathing room + exitmarkdown## Room: [ID] — [Name] **Dimensions**: ~[W]m × [D]m × [H]m **Primary Function**: [Combat / Traversal / Story / Reward] **Cover Objects**: - 2× low cover (waist height) — center cluster - 1× destructible pillar — left flank - 1× elevated position — rear right (accessible via crate stack) **Lighting**: - Primary: warm directional from [direction] — guides eye toward exit - Secondary: cool fill from windows — contrast for readability - Accent: flickering [color] on objective marker **Entry/Exit**: - Entry: [Door type, visibility on entry] - Exit: [Visible from entry? Y/N — if N, why?] **Environmental Story Beat**: [What does this room's prop placement tell the player about the world?]
markdown## Readability Review Critical Path - [ ] Exit visible within 3 seconds of entering room - [ ] Critical path lit brighter than optional paths - [ ] No dead ends that look like exits Combat - [ ] All enemies visible before player enters engagement range - [ ] At least 2 tactical options from entry position - [ ] Fallback position exists and is spatially obvious Exploration - [ ] Optional areas marked by distinct lighting or color - [ ] Reward visible from the choice point (temptation design) - [ ] No navigation ambiguity at junctions
You're successful when:
/hiring and /resource-manager wiring..harness/documents/{mission_name}/workers/{name}.md unless the requester specifies another mission document.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | pass→pass | 13,833 | 16,756 | +21% | 1 | 1 | 0% | 2,042 | 4,646 | +128% | 0 | 0 | — |
case-12 | pass→pass | 16,087 | 16,489 | +2% | 1 | 1 | 0% | 2,472 | 4,792 | +94% | 0 | 0 | — |
case-01 | fail→fail | 47,981 | 26,877 | -44% | 1 | 1 | 0% | 4,506 | 6,327 | +40% | 0 | 0 | — |
case-02 | fail→fail | 22,158 | 16,551 | -25% | 1 | 1 | 0% | 3,501 | 4,837 | +38% | 0 | 0 | — |
case-03 | fail→fail | 18,567 | 22,929 | +23% | 1 | 1 | 0% | 3,295 | 5,738 | +74% | 0 | 0 | — |
case-04 | pass→pass | 11,924 | 14,904 | +25% | 1 | 1 | 0% | 1,706 | 4,325 | +154% | 0 | 0 | — |
case-05 | pass→pass | 13,302 | 9,999 | -25% | 1 | 1 | 0% | 1,968 | 3,675 | +87% | 0 | 0 | — |
case-06 | pass→pass | 15,354 | 25,239 | +64% | 1 | 1 | 0% | 2,065 | 4,470 | +116% | 0 | 0 | — |
case-07 | pass→pass | 15,916 | 13,181 | -17% | 1 | 1 | 0% | 2,401 | 4,213 | +75% | 0 | 0 | — |
case-08 | pass→pass | 10,691 | 23,638 | +121% | 1 | 1 | 0% | 1,650 | 4,357 | +164% | 0 | 0 | — |
case-09 | pass→pass | 13,732 | 9,962 | -27% | 1 | 1 | 0% | 1,941 | 3,770 | +94% | 0 | 0 | — |
case-10 | pass→pass | 10,429 | 10,601 | +2% | 1 | 1 | 0% | 1,594 | 4,040 | +153% | 0 | 0 | — |
case-11 | pass→pass | 15,846 | 27,058 | +71% | 1 | 1 | 0% | 2,287 | 4,581 | +100% | 0 | 0 | — |
case-14 | pass→pass | 16,355 | 13,182 | -19% | 1 | 1 | 0% | 2,328 | 4,136 | +78% | 0 | 0 | — |
case-15 | pass→pass | 13,165 | 11,635 | -12% | 1 | 1 | 0% | 2,081 | 3,951 | +90% | 0 | 0 | — |
case-16 | pass→pass | 17,889 | 19,138 | +7% | 1 | 1 | 0% | 2,643 | 5,113 | +93% | 0 | 0 | — |
case-17 | pass→pass | 17,853 | 16,416 | -8% | 1 | 1 | 0% | 2,726 | 4,699 | +72% | 0 | 0 | — |
case-18 | fail→pass | 13,751 | 6,621 | -52% | 1 | 1 | 0% | 2,166 | 3,269 | +51% | 0 | 0 | — |
case-19 | fail→pass | 13,808 | 11,580 | -16% | 1 | 1 | 0% | 2,078 | 3,945 | +90% | 0 | 0 | — |
case-20 | pass→pass | 11,063 | 13,658 | +23% | 1 | 1 | 0% | 1,546 | 4,206 | +172% | 0 | 0 | — |
case-21 | pass→fail | 19,149 | 19,773 | +3% | 1 | 1 | 0% | 3,436 | 6,007 | +75% | 0 | 0 | — |
case-22 | pass→pass | 19,959 | 20,111 | +1% | 1 | 1 | 0% | 3,794 | 5,924 | +56% | 0 | 0 | — |
case-23 | pass→fail | 26,461 | 22,461 | -15% | 1 | 1 | 0% | 4,127 | 6,050 | +47% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 23 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.