From 4f40f55ab3f64a79b07138d2c7647afb221c0c80 Mon Sep 17 00:00:00 2001
From: truffle <truffleagent@gmail.com>
Date: Sat, 16 May 2026 13:10:29 +0000
Subject: [PATCH] memory: dedup facts within a session by normalized message
 text

Single message matching both correction and preference patterns
produced two facts; identical messages repeated within a session
produced N facts. Track normalized text (lowercase + trim) in a
per-call Set and skip both pattern checks if the key is already
seen. First-match wins on the pattern order.

The Slack-fragment accumulation in #84 is partly cross-session
which this doesn't address, but bounding per-session contribution
to one fact per unique utterance is the floor before any cross-
session check earns its complexity.

Signed-off-by: truffle <truffleagent@gmail.com>
---
 src/memory/__tests__/consolidation.test.ts | 45 ++++++++++++++++++++++
 src/memory/consolidation.ts                |  8 ++++
 2 files changed, 53 insertions(+)

diff --git a/src/memory/__tests__/consolidation.test.ts b/src/memory/__tests__/consolidation.test.ts
index ed088ef7..fe01b7b1 100644
--- a/src/memory/__tests__/consolidation.test.ts
+++ b/src/memory/__tests__/consolidation.test.ts
@@ -135,6 +135,51 @@ describe("consolidateSession", () => {
 		expect(storedFacts.length).toBe(0);
 	});
 
+	test("duplicate identical user messages produce a single fact", async () => {
+		const { memory, storedFacts } = createMockMemory();
+		const data = makeTestSessionData({
+			userMessages: [
+				"Actually, the staging server is on port 3001 not 3000",
+				"Actually, the staging server is on port 3001 not 3000",
+				"Actually, the staging server is on port 3001 not 3000",
+			],
+		});
+
+		const result = await consolidateSession(memory, data);
+
+		expect(result.factsExtracted).toBe(1);
+		expect(storedFacts.length).toBe(1);
+	});
+
+	test("a message matching both correction and preference patterns produces a single fact", async () => {
+		const { memory, storedFacts } = createMockMemory();
+		const data = makeTestSessionData({
+			userMessages: ["No, make sure to use feature branches from now on"],
+		});
+
+		const result = await consolidateSession(memory, data);
+
+		expect(result.factsExtracted).toBe(1);
+		expect(storedFacts.length).toBe(1);
+		expect(storedFacts[0].tags).toContain("correction");
+	});
+
+	test("dedup is case- and whitespace-insensitive within a session", async () => {
+		const { memory, storedFacts } = createMockMemory();
+		const data = makeTestSessionData({
+			userMessages: [
+				"I prefer PRs over direct pushes",
+				"  I PREFER PRs over direct pushes  ",
+				"i prefer PRs over direct pushes",
+			],
+		});
+
+		const result = await consolidateSession(memory, data);
+
+		expect(result.factsExtracted).toBe(1);
+		expect(storedFacts.length).toBe(1);
+	});
+
 	test("episode detail includes tools and files", async () => {
 		const { memory, storedEpisodes } = createMockMemory();
 		const data = makeTestSessionData({
diff --git a/src/memory/consolidation.ts b/src/memory/consolidation.ts
index a6e4cc1e..dc76454c 100644
--- a/src/memory/consolidation.ts
+++ b/src/memory/consolidation.ts
@@ -105,8 +105,13 @@ function calculateImportance(data: SessionData): number {
 function extractFactsFromSession(data: SessionData, episodeId: string): SemanticFact[] {
 	const facts: SemanticFact[] = [];
 	const now = new Date().toISOString();
+	const seenKeys = new Set<string>();
 
 	for (const message of data.userMessages) {
+		const key = message.toLowerCase().trim();
+		if (seenKeys.has(key)) {
+			continue;
+		}
 		const lower = message.toLowerCase();
 
 		if (matchesCorrectionPattern(lower)) {
@@ -125,6 +130,8 @@ function extractFactsFromSession(data: SessionData, episodeId: string): Semantic
 				category: "user_preference",
 				tags: ["correction"],
 			});
+			seenKeys.add(key);
+			continue;
 		}
 
 		if (matchesPreferencePattern(lower)) {
@@ -143,6 +150,7 @@ function extractFactsFromSession(data: SessionData, episodeId: string): Semantic
 				category: "user_preference",
 				tags: ["preference"],
 			});
+			seenKeys.add(key);
 		}
 	}