fix(openai): strip leaked tool result markers

Author: NgoQuocViet2001

internal/httpapi/openai/leaked_output_sanitize_test.go

@@ -35,6 +35,24 @@ func TestSanitizeLeakedOutputRemovesFullwidthDelimitedMetaMarkers(t *testing.T)
 	}
 }
 
+func TestSanitizeLeakedOutputRemovesAssistantEndOfToolCallsMarkers(t *testing.T) {
+	fw := "\uff5c"
+	raw := "A<|Assistant_END_OF_TOOL_CALLS|>B<" + fw + "Assistant▁END▁OF▁TOOL_CALLS" + fw + ">C<|end_of_tool_calls|>D"
+	got := sanitizeLeakedOutput(raw)
+	if got != "ABCD" {
+		t.Fatalf("unexpected sanitize result for assistant end-of-tool-calls markers: %q", got)
+	}
+}
+
+func TestSanitizeLeakedOutputRemovesFullToolResultSection(t *testing.T) {
+	fw := "\uff5c"
+	raw := "开始<" + fw + "Tool" + fw + ">[{\"content\":\"openjdk version 21\"}]<" + fw + "end▁of▁toolresults" + fw + ">结束"
+	got := sanitizeLeakedOutput(raw)
+	if got != "开始结束" {
+		t.Fatalf("unexpected sanitize result for leaked tool result section: %q", got)
+	}
+}
+
 func TestSanitizeLeakedOutputRemovesThinkAndBosMarkers(t *testing.T) {
 	raw := "A<think>B</think>C<|begin▁of▁sentence|>D<| begin_of_sentence |>E<|begin_of_sentence|>F"
 	got := sanitizeLeakedOutput(raw)

internal/httpapi/openai/shared/leaked_output_sanitize.go

@@ -10,6 +10,7 @@ import (
 var emptyJSONFencePattern = regexp.MustCompile("(?is)```json\\s*```")
 var leakedToolCallArrayPattern = regexp.MustCompile(`(?is)\[\{\s*"function"\s*:\s*\{[\s\S]*?\}\s*,\s*"id"\s*:\s*"call[^"]*"\s*,\s*"type"\s*:\s*"function"\s*}\]`)
 var leakedToolResultBlobPattern = regexp.MustCompile(`(?is)<\s*\|\s*tool\s*\|\s*>\s*\{[\s\S]*?"tool_call_id"\s*:\s*"call[^"]*"\s*}`)
+var leakedToolResultSectionPattern = regexp.MustCompile(`(?is)<[\|\x{ff5c}]\s*tool\s*[\|\x{ff5c}]>[\s\S]*?<[\|\x{ff5c}]\s*end[_▁]of[_▁]tool[_▁]?results\s*[\|\x{ff5c}]>`)
 
 var leakedThinkTagPattern = regexp.MustCompile(`(?is)</?\s*think\s*>`)
 
@@ -29,7 +30,8 @@ var leakedThoughtMarkerPattern = regexp.MustCompile(`(?i)<[\|\x{ff5c}]\s*(?:begi
 // halfwidth or legacy U+FF5C fullwidth delimiters:
 //   - ASCII underscore: <|end_of_sentence|>, <|end_of_toolresults|>, <|end_of_instructions|>
 //   - U+2581 variant:   <|end▁of▁sentence|>, <|end▁of▁toolresults|>, <|end▁of▁instructions|>
-var leakedMetaMarkerPattern = regexp.MustCompile(`(?i)<[\|\x{ff5c}]\s*(?:assistant|tool|end[_▁]of[_▁]sentence|end[_▁]of[_▁]thinking|end[_▁]of[_▁]thought|end[_▁]of[_▁]toolresults|end[_▁]of[_▁]instructions)\s*[\|\x{ff5c}]>`)
+//   - compound assistant markers: <|Assistant_END_OF_TOOL_CALLS|>
+var leakedMetaMarkerPattern = regexp.MustCompile(`(?i)<[\|\x{ff5c}]\s*(?:assistant(?:[_▁]end[_▁]of[_▁]tool[_▁]?calls)?|tool|end[_▁]of[_▁]sentence|end[_▁]of[_▁]thinking|end[_▁]of[_▁]thought|end[_▁]of[_▁]tool[_▁]?results|end[_▁]of[_▁]tool[_▁]?calls|end[_▁]of[_▁]instructions)\s*[\|\x{ff5c}]>`)
 
 // leakedAgentXMLBlockPatterns catch agent-style XML blocks that leak through
 // when the sieve fails to capture them. These are applied only to complete
@@ -52,6 +54,7 @@ func sanitizeLeakedOutput(text string) string {
 	}
 	out := emptyJSONFencePattern.ReplaceAllString(text, "")
 	out = leakedToolCallArrayPattern.ReplaceAllString(out, "")
+	out = leakedToolResultSectionPattern.ReplaceAllString(out, "")
 	out = leakedToolResultBlobPattern.ReplaceAllString(out, "")
 	out = stripDanglingThinkSuffix(out)
 	out = leakedThinkTagPattern.ReplaceAllString(out, "")