본문으로 건너뛰기
cccskills
무료GitHub에 공개

tune-enemy-ai

Build and tune enemy behavior for action games, including targeting, movement, attacks, retreats, and boss phases. Organize decisions into explicit states, make attacks readable to players, and check behavior with repeatable tests and browser encounters.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

원문원문 설명 보기

Build, debug, balance, or test combat enemy AI for playable action games. Use for aggro, target selection, navigation, spacing, attack choices, telegraphs, retreats, boss behavior, behavior-state machines, and deterministic AI regression tests.

설치 방법 보기

이럴 때 유용합니다

  • Implementing enemy detection, pursuit, and retreat
  • Tuning attack warnings and combat spacing
  • Debugging boss phases and interrupted actions
  • Testing behavior after target loss or path failure

포함된 파일(2)

  • SKILL.md1.7 KB
  • agents/openai.yaml195 B

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Tune Enemy AI

Make enemy choices legible, bounded, and reproducible.

Model decisions explicitly

Use a small state machine or utility layer with named states such as idle, investigate, pursue, reposition, windup, attack, recover, stagger, retreat, and defeated. State transitions must state their prerequisites, exit conditions, minimum dwell time, and cooldown effects.

Separate perception, intent, and motion

  1. Gather observable inputs: distance, line of sight, target state, occupancy, threat, health, and timers.
  2. Select one intention from constrained legal actions.
  3. Move and animate toward that intent without rewriting the decision mid-action.

Use authoritative collision and navigation results for movement success. Do not derive them from rendered pose or assumed path completion.

Preserve fair combat

Telegraph attacks before their active window. Prevent instant turn-and-hit behavior, perpetual chase, clipped attacks through blockers, and repeated recovery spam. Add spacing and commitment so the player can read and answer each enemy archetype.

Test the decision surface

Create deterministic fixtures for target acquisition, target loss, obstruction, path failure, close-range pressure, multiple enemies, retaliation, interrupt, stagger, boss phase, and reset. Assert transitions and outcomes, not only final positions. Run a real browser encounter after automated tests.

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Build woven fabric in Three.js with visible threads, realistic sheen, seams, and thick hems. The skill covers folds that preserve texture proportions, light passing through thin cloth, and variations in dye and wear.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Creating backlit curtains
  • Differentiating linen, silk, and wool
  • Adding hems and stitches for close-up views
MengTo/Skills6.9천2026년 10월 18일 업데이트

Build falling leaves in Three.js and similar 3D scenes, with visible front, edge, and back views as they tumble and drift. Coordinate wind, scene depth, camera-aware recycling, and performance checks for a convincing autumn effect.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Adding falling autumn leaves to a 3D scene
  • Coordinating leaf tumble, sideways drift, and wind
  • Managing leaves around buildings and camera movement
MengTo/Skills6.9천2026년 10월 18일 업데이트

Add coordinated seasonal transitions to an existing 3D scene. Blend foliage, lighting, sky, ground, snow, and particles from one shared state while preserving camera controls and interactions, so the same place remains recognizable throughout spring, summer, fall, and winter.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Add a four-season selector to an outdoor 3D scene
  • Show seasonal changes in an architectural walkthrough
  • Coordinate foliage, sunlight, sky, and ground appearance
MengTo/Skills6.9천2026년 10월 18일 업데이트

Create or integrate detailed 3D models with smooth outlines and believable edges for close-up views. Prepare lower-detail versions for distant viewing, prioritize loading, and measure rendering costs in the actual scene.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Creating a detailed 3D model for close-up views
  • Replacing visibly faceted curves with smoother geometry
  • Preparing detailed architectural assets for rendering
MengTo/Skills6.9천2026년 10월 18일 업데이트

Prepare detailed 3D materials that remain clear in close-up views. The skill guides texture resolution, physically based material maps, surface mapping, filtering, and progressive loading while checking download size and GPU memory use.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Prepare wood, stone, and fabric materials for close-up views
  • Fix texture seams, grain direction, and normal map orientation
  • Preserve floor and foliage detail at oblique or distant views
MengTo/Skills6.9천2026년 10월 18일 업데이트

Create realistic metal surfaces in Three.js by combining structured reflections, studio lighting, scratches, fingerprints, and beveled edges. Includes polished, brushed, and aged finishes for objects such as tools, rings, and metal product parts.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

  • Fixing metal that looks like gray plastic
  • Creating polished or brushed metal finishes
  • Adding patina and worn edges
MengTo/Skills6.9천2026년 10월 18일 업데이트

MengTo의 스킬 모두 보기

이 스킬의 문제 신고