From 5694fefca0130c2d51e85bed3354c6b164d6dd1a Mon Sep 17 00:00:00 2001
From: Your Name <your.email@example.com>
Date: Thu, 19 Feb 2026 01:39:18 +0800
Subject: [PATCH] fix(agent): downscale vision images to 500px base64

---
 lib/ai/auto-agent-graph.ts | 84 ++++++++++++++++++++++++++++++++++++--
 1 file changed, 81 insertions(+), 3 deletions(-)

diff --git a/lib/ai/auto-agent-graph.ts b/lib/ai/auto-agent-graph.ts
index 2fae37353..06308904a 100644
--- a/lib/ai/auto-agent-graph.ts
+++ b/lib/ai/auto-agent-graph.ts
@@ -1,5 +1,6 @@
 import { Annotation, StateGraph, START, END } from "@langchain/langgraph";
 import { generateText } from "ai";
+import sharp from "sharp";
 import type { AnalysisResult } from "@/lib/ai/smart-analyzer";
 import type { RouteResult } from "@/lib/ai/model-router";
 import { isAiModelMultimodal } from "@/lib/db/ai-model-flags";
@@ -84,6 +85,83 @@ function compactText(text: string, max = 48) {
   return `${normalized.slice(0, max)}...`;
 }
 
+const AGENT_VISION_MAX_IMAGES = 4;
+const AGENT_VISION_MAX_SIDE = 500;
+const AGENT_VISION_FETCH_TIMEOUT_MS = 10_000;
+
+function isHttpUrl(value: string) {
+  return /^https?:\/\//i.test(value);
+}
+
+async function fetchWithTimeout(url: string, timeoutMs: number): Promise<Response> {
+  const controller = new AbortController();
+  const timeoutId = setTimeout(() => controller.abort(), timeoutMs);
+  try {
+    return await fetch(url, {
+      signal: controller.signal,
+      cache: "no-store",
+      headers: {
+        "user-agent": "ai-chatbot-auto-agent/1.0",
+      },
+    });
+  } finally {
+    clearTimeout(timeoutId);
+  }
+}
+
+async function toAgentVisionDataUrl(imageUrl: string): Promise<string | null> {
+  try {
+    const response = await fetchWithTimeout(imageUrl, AGENT_VISION_FETCH_TIMEOUT_MS);
+    if (!response.ok) {
+      console.warn(
+        `[Auto Agent] Failed to fetch image for inline vision payload (${response.status}): ${imageUrl}`,
+      );
+      return null;
+    }
+
+    const contentType = response.headers.get("content-type") || "";
+    if (!contentType.startsWith("image/")) {
+      console.warn(`[Auto Agent] Skip non-image content-type for vision payload: ${contentType}`);
+      return null;
+    }
+
+    const arrayBuffer = await response.arrayBuffer();
+    const optimized = await sharp(Buffer.from(arrayBuffer))
+      .rotate()
+      .resize({
+        width: AGENT_VISION_MAX_SIDE,
+        height: AGENT_VISION_MAX_SIDE,
+        fit: "inside",
+        withoutEnlargement: true,
+      })
+      .jpeg({
+        quality: 80,
+        mozjpeg: true,
+      })
+      .toBuffer();
+
+    return `data:image/jpeg;base64,${optimized.toString("base64")}`;
+  } catch (error) {
+    console.warn("[Auto Agent] Failed to convert image URL to inline base64:", error);
+    return null;
+  }
+}
+
+async function buildAgentVisionInputs(imageUrls: string[]) {
+  if (imageUrls.length === 0) return [] as Array<{ type: "image"; image: string }>;
+
+  const convertedResults = await Promise.all(imageUrls.map((url) => toAgentVisionDataUrl(url)));
+  const inlineCount = convertedResults.filter(Boolean).length;
+  console.log(
+    `[Auto Agent] Prepared vision images: total=${imageUrls.length}, inline=${inlineCount}, remote=${imageUrls.length - inlineCount}, maxSide=${AGENT_VISION_MAX_SIDE}`,
+  );
+
+  return imageUrls.map((originalUrl, index) => ({
+    type: "image" as const,
+    image: convertedResults[index] || originalUrl,
+  }));
+}
+
 function normalizeAspectRatio(value?: string): string | undefined {
   if (!value) return undefined;
   const normalized = value.trim().toLowerCase();
@@ -517,8 +595,8 @@ async function generateAgentConversation(
       .join("\n");
 
     const validImageUrls = state.imageUrls
-      .filter((url) => /^https?:\/\//i.test(url))
-      .slice(0, 4);
+      .filter((url) => isHttpUrl(url))
+      .slice(0, AGENT_VISION_MAX_IMAGES);
 
     const imageInfo = state.hasImage
       ? `用户本轮可用图片数量：${state.imageUrls.length || 1}。图片来源：${state.imageFromHistory ? "历史消息复用" : "本轮上传"}。`
@@ -542,7 +620,7 @@ async function generateAgentConversation(
 2) 只写客观可见信息，不要编造品牌、尺寸、镜头参数。
 3) 全部输出控制在 6 行内。`;
 
-        const visionInputs = validImageUrls.map((url) => ({ type: "image" as const, image: url }));
+        const visionInputs = await buildAgentVisionInputs(validImageUrls);
         const { text: visionText } = await generateText({
           model: visionModel,
           messages: [