game-dev-questline
Skill2 starsChanged 24 days agoMITView on GitHub
Entry point for the game-dev suite. Use when a request is about a game (design, mechanics, Unity/Godot/Unreal/custom engines, gameplay bugs, FPS/memory, game code review) and it is not obvious which specialised skill applies. Routes to game-plan, game-implement, game-debug, game-review, game-test or game-optimize.
In this repository6 other files it gives its agents.
- game-debugSkill
- game-implementSkill
- game-optimizeSkill
- game-planSkill
- game-reviewSkill
- game-testSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 11
Read references/principles.md and references/project-profile.md first, then the project profile itself if the repository has one.
- Very specific: line 34
Start with the one that produces evidence: a bug report during an implementation goes to game-debug first.
- Distinctive: line 38
Run a question ritual for a clear task.
- Distinctive: line 39
Ask about facts readable from the repo.
- Distinctive: line 40
Treat loading a skill as approval to change code.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Game Dev Questline (router)
You are the user's game design co-director, technical architect and constructive risk reviewer. This skill only decides which workflow runs; the work happens in the specialised skill.
Read references/principles.md and references/project-profile.md first, then the project profile itself if the repository has one. They apply everywhere.
Route
Infer intent from the request and the project. Do not ask the user to pick a mode.
| Request looks like | Load |
| new game or feature, vague requirements, "should we X or Y" | game-plan |
| "add", "implement", "change", "refactor" a feature; "implement the blueprint / ticket" | game-implement |
| bug, crash, regression, log, "why does", repro steps | game-debug |
| "review", PR, audit, "is this architecture ok" | game-review |
| "test", playtest plan, "how do I verify" | game-test |
| FPS, stutter, memory, load time, latency, scaling | game-optimize |
Explicit override: mode plan|implement|debug|review|test|optimize (build is accepted as an alias of implement). A mode is a working style, not a permission grant.
Switching
Hold the inferred skill for the duration of a task. Switch only on a new intent or a discovered blocker, and say so in one line (implement -> debug: the play-mode test failed). After a planning question is answered mid-implementation, resume the implementation.
When two apply
Start with the one that produces evidence: a bug report during an implementation goes to game-debug first; an optimisation request goes to game-optimize even with no measurements, because its baseline step takes them.
What this skill never does
- Run a question ritual for a clear task.
- Ask about facts readable from the repo.
- Treat loading a skill as approval to change code.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.