diff --git a/docs/.vitepress/markdownAssets.test.ts b/docs/.vitepress/markdownAssets.test.ts
new file mode 100644
index 0000000..2323d4e
--- /dev/null
+++ b/docs/.vitepress/markdownAssets.test.ts
@@ -0,0 +1,52 @@
+import { existsSync, readdirSync, readFileSync } from "node:fs";
+import { dirname, join, relative, resolve } from "node:path";
+import { fileURLToPath } from "node:url";
+import { describe, expect, it } from "vitest";
+
+const docsRoot = fileURLToPath(new URL("../", import.meta.url));
+const markdownImage = /!\[[^\]]*\]\(([^)]+)\)/g;
+
+const findMarkdownFiles = (directory: string): string[] =>
+  readdirSync(directory, { withFileTypes: true }).flatMap((entry) => {
+    const path = join(directory, entry.name);
+
+    if (entry.isDirectory()) return findMarkdownFiles(path);
+    return entry.isFile() && entry.name.endsWith(".md") ? [path] : [];
+  });
+
+const isExternalAsset = (href: string) =>
+  href.startsWith("#") ||
+  href.startsWith("/") ||
+  href.startsWith("//") ||
+  /^[a-z][a-z\d+.-]*:/i.test(href);
+
+describe("Markdown image assets", () => {
+  it("uses valid POSIX paths that resolve to local files", () => {
+    const invalidAssets: string[] = [];
+
+    for (const markdownFile of findMarkdownFiles(docsRoot)) {
+      const content = readFileSync(markdownFile, "utf8");
+
+      for (const match of content.matchAll(markdownImage)) {
+        const href = match[1].trim();
+        if (isExternalAsset(href)) continue;
+
+        const source = `${relative(docsRoot, markdownFile)} → ${href}`;
+        if (href.includes("\\")) {
+          invalidAssets.push(`${source} (use / instead of \\)`);
+          continue;
+        }
+
+        const assetPath = resolve(
+          dirname(markdownFile),
+          decodeURIComponent(href),
+        );
+        if (!existsSync(assetPath)) {
+          invalidAssets.push(`${source} (file not found)`);
+        }
+      }
+    }
+
+    expect(invalidAssets).toEqual([]);
+  });
+});
diff --git a/docs/cases/submissions/wechat-ima-knowledge/index.md b/docs/cases/submissions/wechat-ima-knowledge/index.md
index 44a875e..ace1a1b 100644
--- a/docs/cases/submissions/wechat-ima-knowledge/index.md
+++ b/docs/cases/submissions/wechat-ima-knowledge/index.md
@@ -76,13 +76,13 @@ ima 知识库连接器是 WorkBuddy 与 ima 之间的桥梁——WorkBuddy 可
 - **图片/截图**：长按图片 → 用小程序打开 → 选择 ima
 
 日积月累，你的 ima 知识库里就有了原始素材。
-![workbuddy添加ima连接器](assets\小程序一键存入ima.jpg)
+![将微信收藏保存到 ima](./assets/小程序一键存入ima.jpg)
 
 ### 第二步：在 WorkBuddy 中添加 ima 连接器
 
 在 WorkBuddy 中打开连接器管理面板，搜索并添加「ima 知识库」连接器，完成授权。授权完成后，WorkBuddy 就能读取你的 ima 知识库内容了。
 
-![workbuddy添加ima连接器](assets\workbuddy添加ima连接器.png)
+![在 WorkBuddy 中添加 ima 知识库连接器](./assets/workbuddy添加ima连接器.png)
 
 
 ### 第三步：用 WorkBuddy 梳理知识体系