diff --git a/scripts/issue-extract.js b/scripts/issue-extract.js
index 874037a..d9a5423 100644
--- a/scripts/issue-extract.js
+++ b/scripts/issue-extract.js
@@ -14,6 +14,7 @@ import {
   expectOk,
   giteaRequest,
   main,
+  pages,
   parseFlags,
   parseRepo,
   preflight,
@@ -92,16 +93,20 @@ async function fetchIssue(login, repo, index, path) {
  *
  * 已整併的留言會被打上 `+1` reaction（由 sdlc-sync 負責標記），而 Gitea 的留言物件
  * 不含 reaction，所以只能逐則再查一次。留言多時請求數會跟著長，但這個數字要準
- * ——它決定下游會不會拿著過期的描述做事。
+ * ——它決定下游會不會拿著過期的描述做事，所以留言也要逐頁讀完，讀不完寧可報錯。
  */
 async function countUnmergedComments(login, repo, commentsPath) {
-  const comments = expectOk(await giteaRequest(login, 'GET', commentsPath), `GET ${commentsPath}`) ?? [];
-
   let unmerged = 0;
-  for (const comment of comments) {
-    const path = `/repos/${repo}/issues/comments/${comment.id}/reactions`;
-    const reactions = expectOk(await giteaRequest(login, 'GET', path), `GET ${path}`) ?? [];
-    if (!reactions.some((reaction) => reaction.content === '+1')) unmerged += 1;
+
+  for await (const comments of pages(login, commentsPath, {
+    limitCode: 'COMMENT_LIMIT',
+    limitHint: `${commentsPath} 的留言太多，數不完未整併的則數`,
+  })) {
+    for (const comment of comments) {
+      const path = `/repos/${repo}/issues/comments/${comment.id}/reactions`;
+      const reactions = expectOk(await giteaRequest(login, 'GET', path), `GET ${path}`) ?? [];
+      if (!reactions.some((reaction) => reaction.content === '+1')) unmerged += 1;
+    }
   }
   return unmerged;
 }