Merge pull request #600 from LHMQ878/fix/555-hallucinated-parameter-suffix

fix: strip trailing mXXXX</parameter> hallucinations

Author: Tarquinen

lib/messages/utils.ts

@@ -7,6 +7,7 @@ const SUMMARY_ID_HASH_LENGTH = 16
 const DCP_BLOCK_ID_TAG_REGEX = /(<dcp-message-id(?=[\s>])[^>]*>)b\d+(<\/dcp-message-id>)/g
 const DCP_PAIRED_TAG_REGEX = /<dcp[^>]*>[\s\S]*?<\/dcp[^>]*>/gi
 const DCP_UNPAIRED_TAG_REGEX = /<\/?dcp[^>]*>/gi
+const HALLUCINATED_PARAMETER_SUFFIX_REGEX = /\nm\d+<\/parameter>\s*$/
 
 const generateStableId = (prefix: string, seed: string): string => {
     const hash = createHash("sha256").update(seed).digest("hex").slice(0, SUMMARY_ID_HASH_LENGTH)
@@ -163,7 +164,8 @@ export const replaceBlockIdsWithBlocked = (text: string): string => {
 }
 
 export const stripHallucinationsFromString = (text: string): string => {
-    return text.replace(DCP_PAIRED_TAG_REGEX, "").replace(DCP_UNPAIRED_TAG_REGEX, "")
+    const withoutHallucinatedParameter = text.replace(HALLUCINATED_PARAMETER_SUFFIX_REGEX, "")
+    return withoutHallucinatedParameter.replace(DCP_PAIRED_TAG_REGEX, "").replace(DCP_UNPAIRED_TAG_REGEX, "")
 }
 
 export const stripHallucinations = (messages: WithParts[]): void => {

tests/message-priority.test.ts

@@ -814,6 +814,13 @@ test("hallucination stripping does not affect non-dcp tags", async () => {
     )
 })
 
+test("hallucination stripping removes trailing mXXXX</parameter> artifact (issue #555)", () => {
+    assert.equal(
+        stripHallucinationsFromString("Total: maybe 20 lines changed.\n\nm0340</parameter>\n\n"),
+        "Total: maybe 20 lines changed.\n\n",
+    )
+})
+
 test("injectMessageIds skips empty assistant messages to avoid prefill (issue #463)", () => {
     const sessionID = "ses_empty_assistant"
     const messages: WithParts[] = [