diff --git a/.jules/bolt.md b/.jules/bolt.md
index d3d39015..5ac660fb 100644
--- a/.jules/bolt.md
+++ b/.jules/bolt.md
@@ -24,3 +24,6 @@
 ## 2026-06-29 - Optimize D1 Queries by Batching
 **Learning:** Found N+1 query pattern on the hub stats page (`hub/src/routes/admin/stats.ts`) when iterating over the list of inbound peers to query `events_pulled_24h` for each peer sequentially using `.first()`. This introduces significant roundtrip latency and resource overhead as the number of peers grows. D1 supports batching queries, which can collapse N queries into a single HTTP roundtrip. D1's returned `batch` types can be peculiar, particularly when running against tests using Miniflare or test shims where row result structure might differ across platforms.
 **Action:** Replaced sequential queries in `Promise.all` with `db.batch()` to combine the multiple `db.prepare(...).bind(...)` calls into a single execution, and normalized the output structure parsing. This significantly reduces N+1 overhead and improves performance scaling as the peer count increases.
+## 2026-07-28 - Optimize N+1 Query in Agent Thread Context Parsing
+**Learning:** In `workers/agent/index.ts`, when preparing the thread context to check for prompt injections and auto-trigger agent workflows, the codebase was mapping over `threadEmails` (retrieved via `stub.getEmails`) and calling `stub.getEmail` inside a `Promise.all` for each email in the thread to get the full body. This caused an N+1 query issue for large threads, generating many separate roundtrips to the Durable Object.
+**Action:** Replaced the `Promise.all` loop with a single call to the existing helper function `getFullThread(stub, emailData.threadId)`. This leverages the DO `getThreadEmails` RPC method, which runs a single combined DO SQL query (batching the fetches), significantly reducing latency and compute overhead for deep conversation threads.
diff --git a/workers/agent/index.ts b/workers/agent/index.ts
index a651a819..e0421adc 100644
--- a/workers/agent/index.ts
+++ b/workers/agent/index.ts
@@ -18,6 +18,7 @@ import {
 } from "../../shared/folders";
 import { isPromptInjection, verifyDraft } from "../lib/ai";
 import {
+	getFullThread,
 	getMailboxStub,
 	stripHtmlToText,
 	textToHtml,
@@ -437,24 +438,10 @@ export class EmailAgent extends AIChatAgent<any> {
 				thread_id: emailData.threadId,
 			})) as EmailMetadata[];
 			if (threadEmails.length > 1) {
-				const fullThread = await Promise.all(
-					threadEmails.map(async (e) => {
-						const full = (await stub.getEmail(e.id)) as EmailFull | null;
-						const text = full?.body ? stripHtmlToText(full.body) : "";
-						return {
-							id: e.id,
-							sender: e.sender,
-							recipient: e.recipient,
-							subject: e.subject,
-							date: e.date,
-							folder_id: e.folder_id,
-							body_text: text,
-						};
-					}),
-				);
-				// ⚡ Bolt: Optimize sorting by replacing `new Date(b.date).getTime()` with `Date.parse(b.date)`
-				fullThread.sort((a, b) => Date.parse(a.date) - Date.parse(b.date));
-				threadContext = fullThread
+				// ⚡ Bolt: Eliminate N+1 query overhead by replacing Promise.all mapping of `stub.getEmail`
+				// with a single call to `getFullThread`, which fetches the entire thread context in one DO roundtrip.
+				const fullThreadData = await getFullThread(stub, emailData.threadId);
+				threadContext = fullThreadData.messages
 					.map(
 						(e) =>
 							`[${e.date}] ${e.sender} → ${e.recipient} (${e.folder_id}): ${e.body_text.substring(0, 500)}`,