
Tune Enemy AI
Design, debug, and test combat enemy behavior for action games with clear, fair states
- Repository
- MengTo/Skills
- Stars
- 7,187
- Licence
- MIT
- Files
- 2 files
What it does
This skill gives an AI coding agent a playbook for building the brain of video game enemies. Once installed, the agent structures enemy behavior as named states (such as idle, pursue, windup, attack, retreat), sets rules for when and how each state can change, and makes attacks telegraphed so players can see them coming. It also guides the agent to write deterministic tests that check enemy decisions and transitions, and to run a real browser encounter to verify the result.
Great for
- Indie or hobby game developers building combat enemies or bosses in an action game
- Game programmers who want a consistent structure for enemy state machines and balancing
- Teams adding automated regression tests to existing enemy AI
Skip it if
- Non-game software, since the guidance is specific to real-time combat enemies (not a general AI skill)
- Projects needing a finished, ready-made enemy; the skill guides design and testing but does not supply game assets or an engine integration
Try asking
- “Build a state machine for a melee grunt enemy with idle, pursue, windup, attack, and recover states”
- “Debug why my boss keeps chasing the player forever and never attacks”
- “Add deterministic tests for target loss, path failure, and stagger in my enemy AI”
- “Make the enemy telegraph its attacks so the player has time to dodge”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 2 original files, the MIT licence and an attribution note.
tune-enemy-ai.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip tune-enemy-ai.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip tune-enemy-ai.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Name your game engine and language in the request so the agent can adapt the guidance to your codebase.
- Ask for the decision logic and tests separately from rendering or animation code to keep the design clean.
- After the agent writes tests, ask it to run a real browser or game encounter as the skill recommends, since automated tests alone may miss feel problems.
What's inside
- agents/openai.yaml
- SKILL.md
Redistributed unmodified from MengTo/Skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.