feat(agent): 统一专业拆镜入口与真实镜头生成
This commit is contained in:
王志伟
@@ -15,7 +15,7 @@ import { runBackendToolGenerationTask } from "@/services/api/generation-task";
|
||||
import { inspectAgentImage } from "@/services/agent-image-preview";
|
||||
import { AGENT_CAPABILITY_GUIDANCE, buildAgentPluginOperations, listAgentCapabilities, readAgentPluginDocumentation, readAgentPluginNode } from "@/services/agent-capabilities";
|
||||
import { getNodeDefinition } from "@/lib/canvas/node-registry";
|
||||
import { buildAgentStoryboardOperations, readAgentStoryboard, validateAgentStoryboardCreation } from "@/lib/canvas/canvas-agent-storyboard";
|
||||
import { buildAgentStoryboardDraft, buildAgentStoryboardOperations, readAgentStoryboard, validateAgentStoryboardCreation } from "@/lib/canvas/canvas-agent-storyboard";
|
||||
import { canvasStylePresets, userStylePreset, type CanvasStylePreset } from "./canvas-style-picker-modal";
|
||||
import { listStyleProfiles } from "@/services/api/style-profiles";
|
||||
import { getActiveUserScope } from "@/lib/user-scope";
|
||||
@@ -137,6 +137,8 @@ function generationToolDefinition(name: string, description: string, mode?: "tex
|
||||
|
||||
const ONLINE_AGENT_TOOLS: ResponseFunctionTool[] = [
|
||||
toolDefinition("canvas_start_art_critique", "打开真实审美节点并准备分析报价，费用卡确认后才提交模型；不会由本工具批准费用。已有当前图片报告直接复用，只有用户明确要求重新分析才传 retry=true。", { nodeId: { type: "string" }, retry: { type: "boolean" } }, ["nodeId"]),
|
||||
toolDefinition("canvas_create_storyboard", "创建待拆镜的专业分镜节点，将原始剧本放入底部输入框。此时表格为空，不能声称镜头已完成；随后调用 canvas_generate_storyboard。已有分镜节点优先复用。", { title: { type: "string" }, prompt: { type: "string" }, x: { type: "number" }, y: { type: "number" } }, ["title", "prompt"]),
|
||||
toolDefinition("canvas_generate_storyboard", "使用分镜节点原有专业拆镜任务，结合项目画风、角色、素材及技能将剧本拆成真实镜头。程序展示费用与整表替换确认，确认后执行并等待结果；取消后不重复发起。局部修改使用 canvas_edit_storyboard。", { nodeId: { type: "string" }, prompt: { type: "string", description: "原始剧本与拆镜要求；省略时使用节点底部剧本" } }, ["nodeId"]),
|
||||
toolDefinition("canvas_read_plugin_node", "读取已启用插件提供的真实节点状态与已有分析结果，不启动新任务。", { nodeId: { type: "string" } }, ["nodeId"]),
|
||||
toolDefinition("canvas_list_styles", "读取真实项目画风目录；source=system 为系统预设，source=user 为当前用户保存的风格。返回 ID 和说明，不能编造 ID。", { source: { type: "string", enum: ["system", "user"] }, query: { type: "string" } }),
|
||||
toolDefinition("canvas_apply_style", "应用从画风目录发现的真实风格，复用页面画风保存逻辑；关联项目时同步项目设置。已有画风且无需改变时直接沿用。", { source: { type: "string", enum: ["system", "user"] }, id: { type: "string" } }, ["source", "id"]),
|
||||
@@ -288,6 +290,7 @@ type CanvasAssistantPanelProps = {
|
||||
onApplyOps: (ops?: CanvasAgentOp[], context?: { conversationId?: string; messageId?: string; source?: "online" | "local" }) => Promise<CanvasAgentSnapshot>;
|
||||
onApplyStyle: (preset: CanvasStylePreset) => Promise<void>;
|
||||
onStartArtCritique: (nodeId: string, restart: boolean) => void;
|
||||
onGenerateStoryboard: (nodeId: string, prompt: string, signal?: AbortSignal) => Promise<boolean | undefined>;
|
||||
canUndoOps: boolean;
|
||||
undoOpsCount: number;
|
||||
onUndoOps: () => CanvasAgentSnapshot | null;
|
||||
@@ -381,6 +384,7 @@ export function CanvasAssistantPanel({
|
||||
onApplyOps,
|
||||
onApplyStyle,
|
||||
onStartArtCritique,
|
||||
onGenerateStoryboard,
|
||||
canUndoOps,
|
||||
undoOpsCount,
|
||||
onUndoOps,
|
||||
@@ -866,6 +870,14 @@ export function CanvasAssistantPanel({
|
||||
const currentSkills = skillRuntime.agentToolNames("onlineAgent").has(name) ? (await listAddedSkills()).skills : composerSkills;
|
||||
const skillToolResult = await skillRuntime.executeAgentTool("onlineAgent", name, args, currentSkills);
|
||||
if (skillToolResult) return skillToolResult;
|
||||
if (name === "canvas_generate_storyboard") {
|
||||
const { node, readiness } = readAgentStoryboard(current, requireString(args.nodeId, "nodeId"));
|
||||
if (!readiness.canGenerateStoryboard) throw new Error(readiness.blockingReason || "请先配置项目画风");
|
||||
const prompt = requireString(args.prompt ?? node.metadata?.composerContent, "剧本与拆镜要求");
|
||||
const completed = await onGenerateStoryboard(node.id, prompt, generationConsumerControllerRef.current.signal);
|
||||
if (!completed) return { ok: true, waitForUser: true, message: "分镜生成未完成或已取消，请查看节点提示；未完成不代表已生成，不要自动重复发起。", data: { nodeId: node.id, completed: false } };
|
||||
return { ok: true, message: "专业拆镜任务已完成并写入真实分镜表。请读取分镜表检查时长与镜头内容，再继续后续操作。", data: { nodeId: node.id, completed: true } };
|
||||
}
|
||||
if (name === "canvas_start_art_critique") {
|
||||
const nodeId = requireString(args.nodeId, "nodeId");
|
||||
const node = current.nodes.find((item) => item.id === nodeId && item.type === "ai-art-critique");
|
||||
@@ -1562,6 +1574,7 @@ function parseToolArguments(value: string) {
|
||||
}
|
||||
|
||||
export function onlineToolToOps(name: string, input: Record<string, unknown>, snapshot: CanvasAgentSnapshot, config: AiConfig): CanvasAgentOp[] {
|
||||
if (name === "canvas_create_storyboard") return buildAgentStoryboardDraft({ title: requireString(input.title, "title"), prompt: requireString(input.prompt, "prompt"), x: numberOr(input.x, nextCanvasX(snapshot)), y: numberOr(input.y, 0) });
|
||||
if (name === "canvas_edit_storyboard") return buildAgentStoryboardOperations(snapshot, input);
|
||||
if (name === "canvas_plugin_action") return buildAgentPluginOperations(requireString(input.pluginId, "pluginId"), requireString(input.actionId, "actionId"), recordOptional(input.input) || {}, snapshot);
|
||||
if (name === "canvas_apply_ops") return validateAgentStoryboardCreation(requireOps(input.ops));
|
||||
@@ -1781,6 +1794,8 @@ function previewOnlineToolCalls(calls: ResponseToolCall[], snapshot: CanvasAgent
|
||||
}
|
||||
|
||||
function toolCallLabel(name: string) {
|
||||
if (name === "canvas_create_storyboard") return "创建待拆镜分镜表";
|
||||
if (name === "canvas_generate_storyboard") return "专业拆镜";
|
||||
if (name === "canvas_start_art_critique") return "准备审美分析";
|
||||
if (name === "canvas_read_plugin_node") return "读取插件结果";
|
||||
if (name === "canvas_list_styles") return "查找项目画风";
|
||||
|
||||
@@ -393,7 +393,7 @@ export function CanvasScriptNodeContent({
|
||||
className="thin-scrollbar h-full min-h-0 w-full touch-pan-y resize-none overflow-y-auto overflow-x-hidden overscroll-contain rounded-md border bg-transparent px-3 py-2 text-sm leading-5 outline-none transition placeholder:opacity-45 focus:ring-1"
|
||||
style={{ borderColor: theme.node.stroke, color: theme.node.text, "--tw-ring-color": theme.node.muted } as CSSProperties}
|
||||
value={prompt}
|
||||
placeholder="描述想生成的脚本或视频内容"
|
||||
placeholder="输入剧本或剧情，以及总时长等要求，自动拆分为镜头"
|
||||
onContentSizeChange={resizePrompt}
|
||||
onChange={(value) => {
|
||||
setPrompt(value);
|
||||
@@ -447,11 +447,11 @@ export function CanvasScriptNodeContent({
|
||||
value={shotDuration}
|
||||
disabled={node.metadata?.status === "loading"}
|
||||
options={[
|
||||
{ value: "auto", label: "时长自动" },
|
||||
{ value: "5", label: "5 秒" },
|
||||
{ value: "10", label: "10 秒" },
|
||||
{ value: "15", label: "15 秒" },
|
||||
{ value: "30", label: "30 秒" },
|
||||
{ value: "auto", label: "单镜时长自动" },
|
||||
{ value: "5", label: "每镜 5 秒" },
|
||||
{ value: "10", label: "每镜 10 秒" },
|
||||
{ value: "15", label: "每镜 15 秒" },
|
||||
{ value: "30", label: "每镜 30 秒" },
|
||||
]}
|
||||
popupMatchSelectWidth={false}
|
||||
onChange={onShotDurationChange}
|
||||
|
||||
Reference in New Issue
Block a user