flutter3d_mcp library

Classes

Answer: What a tool call actually did, and the sentence to say about it.
AnsweredCall<A>: One call a ToolTableServer answered, as a watcher is told of it: ToolTableServer.onCall with the server's own answer type, ToolTableServer.onProjectCall with a project tool's ToolResult.
CommandTarget: What a call says it acts on, instead of leaving it to the selection — ux-20. See ModelSession.targetIn for how it is read out of a call's own arguments, and ModelSession.runOn for what it does.
DepthSpan: How far in front of the eye something reaches, nearest and farthest.
EditorMcpServer: A level editor, offered to an agent as a table of tools.
EditorSession: One level document, open, with the editor's own verbs on it.
LevelCamera: Where a picture of the level is taken from: the eye, and the point it looks at.
LevelView: The level as the software rasteriser draws it, and what an agent can learn from the drawing — screenshot and report.
LoopbackMcpServer: A server listening on 127.0.0.1.
McpToolRegistry: MCP tools, merged into the project's servers as <namespace>.<tool>: the slot McpTools fills, host.registry<McpTools>().
McpTools: The McpToolRegistry a project fills: the MCP tools its plugins bring, each under the plugin's namespace.
ModelHttpServer: A ModelMcpServer listening on 127.0.0.1, for ModelHttpServer.start.
ModelMcpServer: A model project, offered to an agent as a table of tools.
ModelSession: One model project, open, with the editor's own verbs on it.
OfferedPrompt: One prompt: what prompts/list offers, and the text prompts/get hands back — ux-44.
OfferedTool<S, A>: One tool: what an agent is offered, and what calling it does to the session.
PieceCover: One piece in front of another, and how much of the other it hides.
PieceReport: What one piece of the level looks like from a camera.
PixelBox: A box of whole pixels on the frame, from the top left, every edge inclusive.
PlaySession: HR5 for an agent: the game the open level belongs to, run, reloaded, stopped and sent the level when it is saved — the editor application's Play, through the same FlutterRun.
ProjectMcpServer: A server offering a project's McpTools and nothing else.
ProjectRoot: The directory a server's file tools are held to, and the one place a path an agent sent becomes a path on the disk.
ProjectTool: One tool a plugin, or the application, added to a project's McpTools.
ResolvedPath: What ProjectRoot.tryResolve answers: the resolved path inside the root, or the sentence that refuses it. Exactly one of the two is set.
SessionToolCall<S>: One call of a server's own tool: a ToolCall with the session the server runs it against.
ToolAudio: A sound a tool answers with: the encoded bytes and their media type (audio/wav, audio/ogg).
ToolCall: One call of a project tool: what an agent passed, under the name it called.
ToolContent: One piece of what a tool answers: words, a picture, a sound, a resource or a link to one, as the protocol's content objects say them.
ToolHints: What a tool does to the world it is called on, as the protocol's annotations say it to a host — which may ask a person before a destructive call, and may call a read-only one without asking.
ToolImage: A picture a tool answers with: the encoded bytes and their media type.
ToolName: The name a tool is published under, area.verb, and what it does to the world it is called on.
ToolOtherContent: Content of a kind this build has no class for, kept as the JSON it came as and written back unchanged.
ToolResource: A resource carried inside the answer: its uri, and its contents as text or as bytes (blob).
ToolResourceLink: A link to a resource the client may read, rather than its contents.
ToolResult: What a tool call answers: content for a model to read, and the machine-readable structuredContent when the tool declares an outputSchema.
ToolSpec: One tool, as tools/list offers it: a name, a sentence, the JSON Schema of its arguments, the JSON Schema of the structuredContent it answers with when it answers one, and its ToolHints.
ToolTableServer<S, A>: A server offering tools over one session, every answer turned into a result by toResult.
ToolText: Words a tool answers with.

Constants

buildTargets → const Set<String>: What flutter build is asked to make, by the word an agent gives it.
describableFamilies → const List<String>: The families describeType can be asked about.
editorAnswerSchema → const Map<String, Object?>: The shape of every editor tool's structuredContent, as the outputSchema each tool declares: whether the call did what it was asked (did, false for a refusal), its sentence, the selection after it in the words level.list uses, and whether a picture came with it.
editorMcpSchemaVersion → const String: The version of this server's tools — names and input schemas — as the initialize result announces it beside editorMcpVersion.
editorMcpVersion → const String: The version this server tells a client it is: the pubspec's.
editorToolNames → const Map<String, ToolName>: What each of editorTools is published as, area.verb, and what it does to the document. The name a tool is written under — the editor command's own word, which EditorCommand.fromJson reads — stays answerable as an alias until 2.0.
modelAnswerSchema → const Map<String, Object?>: The shape of every model tool's structuredContent — what ModelSession.structured writes — as the outputSchema each tool declares.
modelMcpSchemaVersion → const String: The version of this server's tools — names and input schemas — as the initialize result announces it beside modelMcpVersion.
modelMcpUsage → const String: What model_mcp prints for --help and for a wrong argument count.
modelMcpVersion → const String: The version this server tells a client it is: the pubspec's.
modelPrompts → const List<OfferedPrompt>: The advice, offered to every client that asks for it.
modelToolNames → const Map<String, ToolName>: What each of the modeller's tools is published as, area.verb, and what it does to the project. The written name — the model command's own word, which history.batch and the command journal still use — stays an alias until 2.0.
projectMcpSchemaVersion → const String: The version of ProjectMcpServer's own tools, of which there are none.
projectMcpVersion → const String: The version ProjectMcpServer tells a client it is: the pubspec's.
renderModes → const List<String>: Every mode an agent can ask for — mcp-08n's own four, plus selection, which the row does not count because it is not a shading mode. wireframe landed 2026-09-17 as geometry rather than as a line topology; RenderShading.wireframe and wire_overlay.dart both say why.
schemaToolName → const String: The name of the tool every server offers to say what schema versions it speaks.
schemaVersionMetaKey → const String: The _meta key a server's tools, and its initialize result, carry their schema version under.

Properties

auditTool → ModelPictureTool: audit: AssetAudit in words and the asset from all seven views in one picture, with repair to put right what has one right answer.
final
editorTools → List<EditorTool>: Everything this server offers: the ten commands, the six verbs that are about the session rather than about the document, and the two that look at it — view.screenshot and view.report.
no setter
modelTools → List<ModelTool>: Everything this server offers: every command as a tool, and the verbs that are about the session rather than about one edit — list and select, because a program with no screen cannot point at anything; undo/redo; check, save, export and import, because a project has a life outside the commands that shape it; journal, because CommandJournal is worth writing from the one place that runs every command already.
no setter
renderSheetTool → ModelPictureTool: A contact sheet of renderSheetViews — mcp-07n's own tool. Labelled: each quadrant carries the name of the view it is, drawn with tiny_font.dart rather than a widget, since this runs in a plain Dart process with no canvas in it.
final
renderSnapshotTool → ModelPictureTool: pro-rn-04: a snapshot at a size and a supersample a person chooses, through pro-rn-02's own tiled job.
final
renderTool → ModelPictureTool: A picture of the project as it stands.
final

Functions

commandInputSchema(PluginCommandInfo command) → Map<String, Object?>: command's arguments as a tool's input schema: an object with one property per argument hint, each required when it has no default.
deleteMcpSessionFile(File file) → void: Removes the session file writeMcpSessionFile wrote — called once the server it named is no longer listening, so nothing reads a port that has gone stale.
describeType({required String family, String? kind}) → String: What kind takes, or — with no kind — every kind of family at once.
pictureResultOf(Answer answer) → ToolResult: answer as a tool result: the sentence, then the PNG when there is one.
refuseArguments(ToolSpec tool, Map<String, Object?> arguments) → String?: Why arguments are not a call to tool, or null when they are.
resultOf(Answer answer) → ToolResult: answer as a tool result: the sentence, marked as an error when it is a refusal.
serveEditorMcp(List<String> arguments, {List<Flutter3dPlugin> plugins = const <Flutter3dPlugin>[]}) → EditorMcpServer: Opens the level arguments names and serves it over stdio, offering the MCP tools plugins bring beside the editor's own.
serveModelMcp(List<String> arguments, {McpTools? projectTools}) → ModelMcpServer?: Opens the project arguments names and serves it over stdio, offering projectTools beside the modeller's own.
shareCommands(EditorPieces pieces, McpTools tools, EditorSession session) → Registration: Offers every command the plugins added to pieces as a tool of tools, <namespace>.<name>, that runs it on session's document; keeps the offer in step while plugins are switched on and off, until the registration it returns is cancelled.
writeMcpSessionFile(File file, {required int port, required String token}) → void: Where an application's own running LoopbackMcpServer can be found, written beside the application's existing session state.

Typedefs

EditorTool = OfferedTool<EditorSession, Answer>: One tool: what an agent is offered, and what calling it does to the document — see flutter3d_mcp/kit.dart's OfferedTool for why a pair rather than a table and a switch.
GameAsk = Future<GameExtensionReply> Function(String vmService, String method, {Map<String, String> args}): Asks a running game for one of its ext.flutter3d.* extensions — see callGameExtension, which is what a session asks with unless a test hands in its own.
ModelPictureTool = OfferedTool<ModelSession, Answer>: A modeller tool that may answer with a picture: a body over the ModelSession that answers an Answer.
ModelTool = OfferedTool<ModelSession, Answer>: One tool: what an agent is offered, and what calling it does — see flutter3d_mcp/kit.dart's OfferedTool for why a pair rather than a table and a switch.
NewRun = FlutterRun Function(String projectRoot, String? device): Makes a run of projectRoot on device. A parameter of PlaySession so a test hands it a run over a tool of its own making.
ProjectToolRun = FutureOr<ToolResult> Function(ToolCall call): What calling a project tool does: one ToolCall in, a result out.
RunFlutterIn = Future<ProcessResult> Function(List<String> arguments, String workingDirectory): Runs flutter with arguments in workingDirectory to the end — what PlaySession.build builds with. A parameter so a test answers for the tool.

Exceptions / Errors

PathOutsideRootException: A path a file tool was handed that lies outside its ProjectRoot.

flutter3d_mcp library

Classes

Constants

Properties

Functions

Typedefs

Exceptions / Errors

flutter3d_mcp package

flutter3d_mcp library