Skip to main content
Glama

Author an interactive beat in one call

make_choice_scene
DestructiveIdempotent

Builds an RPG Maker MZ event page script from a step list, compiling dialog, choices, branches, loops, battles, shops, and transfers with validated switches, variables, and assets.

Instructions

Write an event's whole script from a step list: dialog, choice branches (each option can carry its own when gate), if over switches, variables, items, gold and buttons, loop with break, battle with win/escape/lose branches, shop, transfers, audio, screen effects, gameOver. This is the tool for a scene rather than a person: pass a cell to place a new event, or an existing eventId and pageIndex to rewrite that page. The compiler produces the indentation and the branch markers (402/403, 411/412, 413, 601-603) the engine reads, so a body cannot land outside the branch it belongs to — the mistake that otherwise shows up as a choice that always takes the first option. Checked before writing: unknown step names, conditions or commands that name a switch, variable or database row the game does not have, audio and face files the project lacks. One transaction, map rendered back.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "mapId": "<mapId>",
      "script": "<script>"
    },
    "name": "make_choice_scene"
  }
}

Input Schema

NameRequiredDescriptionDefault
xNo
yNo
nameNo
whenNoCondition for the page this script goes to
imageNoThe event's graphic. A characterName that is not in img/characters fails the call
mapIdYes
scriptYesThe page's whole script. Each step is an object with exactly one key: `say` (a string, a list of lines, or {lines, speaker?, faceName?, faceIndex?, background?, positionType?}), `scroll` ({lines, fast?, wait?}), `comment`, `choice` ({prompt?, options: [{label, then, when?, lockedMessage?}], cancel?: steps, defaultType?, positionType?, background?}), `if` ({when, then, else?}), `loop` (steps, closed by `break` or `goto`), `break`, `exit`, `label`, `goto`, `switch` ({id, value?}), `selfSwitch` ({letter, value?}), `variable` ({id, set?|add?|multiply?}), `gold` (a number; negative takes), `item`/`weapon`/`armor` ({id, count?, take?}), `heal`, `transfer` ({mapId, x, y, direction?, fade?}), `wait` (frames), `fade` ("out"|"in"), `flash` ({color?, duration?, wait?}), `weather` ({type, power, duration?, wait?}), `se`/`me`/`bgm` (a name or {name, volume?, pitch?, pan?}), `stopBgm`, `stopSe`, `animate` ({target?, animationId, wait?}), `balloon` ({target?, balloonId, wait?}), `moveRoute` ({target?, route: [route steps], repeat?, skippable?, wait?}), `battle` ({troopId, canEscape?, canLose?, win?, escape?, lose?}), `shop` ({goods: [{id, kind?, price?, purchaseOnly?}]}), `menu`, `saveScreen`, `gameOver`, `title`, `commonEvent`, `erase`, `script` (engine code, stored as 355/655) and `raw` (verbatim {code, parameters, indent?}). `script` and `raw` are escape hatches: they come back named in the reply, and a build that leans on them is reporting a shape this layer does not cover yet.
eventIdNoRewrite this event instead of placing a new one
replaceNoWith `name`: rebuild the event of that name instead of adding another (default true)
triggerNo0 action button, 1 player touch, 2 event touch, 3 autorun (fires on entering the map), 4 parallel process. 0 and 1 are checked for a cell the player can stand on; 3 and 4 are not, so an opening cutscene can sit on a wall in the corner the way the editor's own do
pageIndexNoWith eventId: which page to replace (default 0)
priorityTypeNo0 below tiles (default: a zone or a cutscene should not block), 1 same as tiles, 2 above

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.4.2

TDQS

A4.6/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Adds substantial behavior beyond the annotations: pre-write validation (unknown steps, missing switches/variables/rows, missing audio and face files), automatic indentation and branch-marker emission, and the 'one transaction, map rendered back' atomicity guarantee. The 'rewrite that page' wording also discloses the overwrite that destructiveHint=true implies.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose and the place-vs-rewrite decision, then the step catalog, then validation/transaction behavior. The step enumeration is long but each token names a real construct; slightly dense as a single block, but no filler sentences.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 12-param, no-output-schema mutation tool this covers purpose, modes, validation and atomicity, and notes that `script`/`raw` escape hatches 'come back named in the reply'. It stops short of describing the overall response payload or pagination/limits, but is close to complete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema leaves `script` as an opaque array of single-key objects (additionalProperties {}), and the description compensates by enumerating the whole step vocabulary and the shape of key steps (choice options with `when`/`then`, battle win/escape/lose branches, loop closed by break/goto). This is meaning the schema does not carry.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Write an event's whole script from a step list') and enumerates the exact constructs it authors (dialog, choice branches, if, loop, battle, shop, etc.). It also positions itself against siblings by declaring 'the tool for a scene rather than a person', separating it from make_npc/make_chest.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear context for both modes ('pass a cell to place a new event, or an existing eventId and pageIndex to rewrite that page') and differentiates scene-level authoring from person-level authoring. It does not explicitly name nearby alternatives like add_commands/set_commands/place_event, so the routing is implied rather than spelled out.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.