game-brief
Skill0 starsChanged 3 days agoMITView on GitHub
Use when someone is planning a Unity 6 mobile game before generating or coding it or asks what to prepare before code, wants a game brief, pitch, MVP list or pillars, or asks whether a game's scope has crept or whether a new feature still fits.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- wisky3107/agentstheir own version, first seen Oct 5, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- wisky3107/agents★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository20 other files it gives its agents.
- compass-kit/CLAUDE.mdCLAUDE.md
- unity-kit/CLAUDE.mdCLAUDE.md
- backend-new-projectSkill
- cmp-backendSkill
- cmp-code-rulesSkill
- cmp-commitSkill
- cmp-design-rulesSkill
- cmp-design-to-codeSkill
- cmp-detektSkill
- cmp-featureSkill
- cmp-gatesSkill
- cmp-maestroSkill
- cmp-new-projectSkill
- cmp-testingSkill
- cmp-verifySkill
- game-balanceSkill
- game-lookSkill
- game-new-projectSkill
- game-releaseSkill
- game-smokeSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 64
If a field needs a second paragraph, cut it.
- Very specific: line 68
Read-only. The MVP list in design/brief.md is the baseline.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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-brief
Two modes. brief writes one file, design/brief.md, that fits one screen. scope-check compares the project with that file and writes nothing. Pick the mode from the argument or the request; "should we add X" or "has scope crept" means scope-check.
Mode: brief
The output is the filled template below and nothing else: no other documents, no analysis before or after it. Tell the user the file path and the three (guess) markers you are least sure of.
Where. Normally design/brief.md (create design/ if missing). If there is no Unity project yet, write brief.md in the current folder and tell the user to move it to design/brief.md after game-new-project. If design/brief.md already exists, show what would change and ask before overwriting.
Asking. From the request, fill everything you can. Ask only what cannot be inferred, in one short message with at most four questions. If the genre or the camera (2D or 3D) cannot be inferred, ask about it even when the user says "go ahead". Everything else still open becomes a guess marked (guess) in the file so the user can correct it. Do not ask for a field you can reasonably propose.
Template for the brief (at most 60 lines):
# <Working title>
**Pitch:** <one sentence>
## Core loop
- <do> -> <get> -> <why again> (2-4 bullets)
## Player goal and fail
<what the player wants; how a run ends. 1-2 lines, testable>
## MVP (7 items at most)
- <feature without which this is not the game>
## Out of scope (5 items at most)
- <specific thing not being built>
## Build order
1. <the riskiest or most core-fun item first>
## Pillars
1. <pillar, one line>
2. <pillar, one line>
3. <pillar, one line>
- **Anti-pillar:** NOT <something the team might plausibly want>
**For whom / feel:** <player and the feeling, one line>
**Art direction:** <one line>
**Reference game:** <game> - the MVP keeps about <N>% of it: <which part> | none (drop the percentage)
## Mobile
**Orientation:** portrait | landscape
**One-handed:** yes | no
**Session length:** <e.g. 2-3 minutes>
**Revenue:** ads | IAP | none | undecided
**Stores:** Google Play | App Store | bothRules for the fields:
**Orientation:**is exactlyportraitorlandscape, on its own line, never with a(guess)marker, becausegame-smokereads it. A guessed orientation goes in the guesses you report to the user.- A pillar is falsifiable and removes options ("one finger, never two" not "fun controls").
- Each build-order step is something the player can try; the first one proves the loop is fun.
- The reference game percentage is the honest size of the MVP against a shipped game.
- If a field needs a second paragraph, cut it; that is a design document, not a brief.
Mode: scope check
Read-only. The MVP list in design/brief.md is the baseline.
- No
design/brief.md, or a brief with no MVP items: reportNOT ASSESSEDand stop. - Gather what the project now contains:
git logsince the brief's last change (the brief's own commit isgit log -1 --format=%H -- design/brief.md; empty output means it has none, so use all history),TODO/FIXMEcomments, and the feature folders, scenes and scripts present. Features that are already in the game's first generated commit (git rev-list --max-parents=0 HEAD) are the starting point, not work: do not count them. If there is no git history, the feature folders and scripts the template ships are the starting point. If the first commit is not the generated one (squashed history, or the project was added to an existing repo), use the commit the user names or the earliest commit that adds the Unity project. Say which baseline you used in the report. - Map each piece of work to the brief. An item is one player-visible feature or system; several commits or files for the same feature count once. A TODO counts only if it names new scope. Work that serves an MVP item without adding a new player-visible capability is not added. Count added: items in neither the MVP nor a reasonable part of it, or that touch an Out of scope item. Count removed: MVP items dropped or deferred. An MVP item that files cannot prove (for example a timing or pacing rule) is listed as "not verifiable" and counts as neither present nor removed.
- Scope change = (added + removed) / number of MVP items in the brief.
| Scope change | Verdict |
| 10% or less | PASS |
| above 10%, up to 25% | CONCERNS |
| above 25% | FAIL |
Report: verdict, the baseline count, each added and removed item with the commit or file that shows it, and one recommendation per added item (cut, defer, or swap for an MVP item). If the user wants the scope change to stay, the brief is edited with their say-so, then re-checked.
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.

