From 5dc1a340391da590b4732cfedc5ff87dd13001ec Mon Sep 17 00:00:00 2001
From: fisker Cheung <lionkay@gmail.com>
Date: Thu, 24 Apr 2025 17:04:42 +0800
Subject: [PATCH] Website: generate `llms-full.txt` and `llms.txt` files
 (#17388)

---
 cspell.json                         |   1 +
 website/docusaurus.config.js        |   3 +-
 website/plugins/llms-txt-plugin.mjs | 100 ++++++++++++++++++++++++++++
 3 files changed, 103 insertions(+), 1 deletion(-)
 create mode 100644 website/plugins/llms-txt-plugin.mjs

diff --git a/cspell.json b/cspell.json
index d9c898528..246432cdd 100644
--- a/cspell.json
+++ b/cspell.json
@@ -159,6 +159,7 @@
         "lintstaged",
         "lintstagedrc",
         "literalline",
+        "llms",
         "loglevel",
         "lydell",
         "Marek",
diff --git a/website/docusaurus.config.js b/website/docusaurus.config.js
index 30d627db5..7fb3f039f 100644
--- a/website/docusaurus.config.js
+++ b/website/docusaurus.config.js
@@ -4,6 +4,7 @@ import fs from "node:fs";
 import { createRequire } from "node:module";
 import { load as parseYaml } from "js-yaml";
 import { themes as prismThemes } from "prism-react-renderer";
+import llmsTxtPlugin from "./plugins/llms-txt-plugin.mjs";
 
 const require = createRequire(import.meta.url);
 
@@ -258,7 +259,7 @@ const config = {
         darkTheme: prismThemes.dracula,
       },
     },
-
+  plugins: [llmsTxtPlugin],
   future: {
     experimental_faster: {
       swcJsLoader: true,
diff --git a/website/plugins/llms-txt-plugin.mjs b/website/plugins/llms-txt-plugin.mjs
new file mode 100644
index 000000000..dd60b5c9c
--- /dev/null
+++ b/website/plugins/llms-txt-plugin.mjs
@@ -0,0 +1,100 @@
+// https://github.com/facebook/docusaurus/issues/10899
+// https://github.com/defi-wonderland/handbook/blob/dev/plugins/llmsTxtPlugin.ts
+import fs from "node:fs/promises";
+import path from "node:path";
+
+export default function llmsTxtPlugin(context) {
+  return {
+    name: "llms-txt-plugin",
+    async loadContent() {
+      const { siteDir } = context;
+      const contentDir = path.join(siteDir, "versioned_docs/version-stable");
+      const allMdx = [];
+      const docsRecords = [];
+
+      const getMdxFiles = async (dir, relativePath = "", depth = 0) => {
+        const entries = await fs.readdir(dir, {
+          withFileTypes: true,
+        });
+
+        entries.sort((a, b) => {
+          if (a.isDirectory() && !b.isDirectory()) {
+            return -1;
+          }
+          if (!a.isDirectory() && b.isDirectory()) {
+            return 1;
+          }
+          return a.name.localeCompare(b.name);
+        });
+
+        for (const entry of entries) {
+          const fullPath = path.join(dir, entry.name);
+          const currentRelativePath = path.join(relativePath, entry.name);
+
+          if (entry.isDirectory()) {
+            const dirName = entry.name
+              .split("-")
+              .map((word) => word.charAt(0).toUpperCase() + word.slice(1))
+              .join(" ");
+            const headingLevel = "#".repeat(depth + 2);
+            allMdx.push(`\n${headingLevel} ${dirName}\n`);
+            await getMdxFiles(fullPath, currentRelativePath, depth + 1);
+          } else if (entry.name.endsWith(".md")) {
+            const content = await fs.readFile(fullPath, "utf8");
+            let title = entry.name.replace(".md", "");
+            let description = "";
+
+            const frontmatterMatch = content.match(/^---\n(.*?)\n---/su);
+            if (frontmatterMatch) {
+              const titleMatch = frontmatterMatch[1].match(/title:\s*(.+)/u);
+              const descriptionMatch =
+                frontmatterMatch[1].match(/description:\s*(.+)/u);
+
+              if (titleMatch) {
+                title = titleMatch[1].trim();
+              }
+              if (descriptionMatch) {
+                description = descriptionMatch[1].trim();
+              }
+            }
+
+            const headingLevel = "#".repeat(depth + 3);
+            allMdx.push(`\n${headingLevel} ${title}\n\n${content}`);
+
+            // Add to docs records for llms.txt
+            docsRecords.push({
+              title,
+              path: currentRelativePath.replaceAll("\\", "/"),
+              description,
+            });
+          }
+        }
+      };
+
+      await getMdxFiles(contentDir);
+      return { allMdx, docsRecords };
+    },
+    async postBuild({ content, outDir }) {
+      const { allMdx, docsRecords } = content;
+
+      // Write concatenated MDX content
+      const concatenatedPath = path.join(outDir, "llms-full.txt");
+      await fs.writeFile(concatenatedPath, allMdx.join("\n\n---\n\n"));
+
+      // Create llms.txt with the requested format
+      const llmsTxt = `# ${
+        context.siteConfig.title
+      }\n\n## Documentation\n\n${docsRecords
+        .map(
+          (doc) =>
+            `- [${doc.title}](${context.siteConfig.url}/docs/${doc.path.replace(
+              ".md",
+              "",
+            )}): ${doc.description}`,
+        )
+        .join("\n")}`;
+      const llmsTxtPath = path.join(outDir, "llms.txt");
+      await fs.writeFile(llmsTxtPath, llmsTxt);
+    },
+  };
+}