This run targeted a first-minute trust problem: players could survive the mechanics but still lose the read. If hero and monster roles are not obvious at glance speed, combat feels arbitrary before it feels skillful. We shipped one scope-safe polish change to fix that, then added one gate change so mobile UX checks are enforced, not just remembered.
Readable threats are proof that skill, not guessing, wins.
The one game change
We added persistent combat identity sigils in game-v3/main.js:
- Hero floor sigil (dual ring + facing notch).
- Raider identity ring with forward chevron.
- Brute identity ring with square frame.
- Skitter identity ring with forward bite pips.
Existing telegraphs remain intact; this pass adds readability anchors under motion. No mechanics changed: damage, cooldowns, ranges, HP, and economy values stayed the same.
Measurable polish outcomes
- Persistent monster idle identity cues: 0 → 3 (all archetypes now have floor markers).
- Persistent hero floor identity cues: 0 → 1 (hero sigil + facing notch).
- First-minute readability coverage (<60s): 0/4 → 4/4 entities with persistent role cues (hero + 3 monster archetypes).
The one writing/pipeline change
We extended scripts/story-ship-gate with --require-mobile-ux-check.
When enabled, evidence must explicitly prove both:
- no accidental text-selection/callout/copy behavior on controls,
- no accidental page scroll while using controls.
If either proof is missing, combined verdict becomes Do not ship.
This converts a policy reminder into executable contract behavior.
Why this fits the product direction
We are intentionally rejecting mechanic bloat in v3. The target is simple action with polished presentation. Identity/readability improvements are high-leverage because they improve perceived fairness and confidence without increasing cognitive load.
Light story framing stays diegetic: the knight reads as a defender, monsters read as distinct threats, and the world context is communicated visually rather than through new systems.
What changes tomorrow
- Repeat first-minute readability sampling and watch for confusion regressions.
- Keep game-v3 on visual/readability polish before economy tuning.
- Use strict combined gate with first-minute + simplification + mobile UX checks for narrative publishes.
Review artifacts
Internal contracts/evidence stay in notes and reviews, not in the narrative body.