Skip to content

Agent Zero AI framework

Fix compaction backups for malformed Unicode

Alessandro98e51fbb2a3f
diff --git a/plugins/_chat_compaction/AGENTS.md b/plugins/_chat_compaction/AGENTS.md
index c1d7b418a..88a4f8d45 100644
--- a/plugins/_chat_compaction/AGENTS.md
+++ b/plugins/_chat_compaction/AGENTS.md
@@ -16,6 +16,7 @@
## Local Contracts
- Preserve chat history integrity and persistence after compaction.
+- Backup JSON and transcript artifacts must remain UTF-8 writable when chat content contains malformed Unicode such as lone surrogates.
- Keep generated summaries bounded by configured model and token limits.
- Do not discard original context data unless the compaction flow explicitly owns that behavior.
diff --git a/plugins/_chat_compaction/helpers/compactor.py b/plugins/_chat_compaction/helpers/compactor.py
index 8561dc620..b61cd7eed 100644
--- a/plugins/_chat_compaction/helpers/compactor.py
+++ b/plugins/_chat_compaction/helpers/compactor.py
@@ -4,7 +4,7 @@ from collections import deque
import models as models_module
from agent import Agent
-from helpers import tokens
+from helpers import files, tokens
from helpers.history import History, output_text
from helpers.persist_chat import (
export_json_chat,
@@ -42,11 +42,8 @@ def _save_pre_compaction_backup(context, full_text: str) -> dict[str, str]:
txt_path = os.path.join(backup_dir, f"pre-compact-{timestamp}.txt")
json_content = export_json_chat(context)
- with open(json_path, "w", encoding="utf-8") as f:
- f.write(json_content)
-
- with open(txt_path, "w", encoding="utf-8") as f:
- f.write(full_text)
+ files.write_file(json_path, json_content)
+ files.write_file(txt_path, full_text)
return {"json": json_path, "txt": txt_path}
diff --git a/tests/test_chat_compaction.py b/tests/test_chat_compaction.py
index 39c4169d7..fa8358c57 100644
--- a/tests/test_chat_compaction.py
+++ b/tests/test_chat_compaction.py
@@ -1,4 +1,5 @@
import sys
+from types import SimpleNamespace
from pathlib import Path
import pytest
@@ -42,6 +43,32 @@ class _RecordingModel:
return f"summary-{len(self.user_messages)}", None
+def test_pre_compaction_backup_sanitizes_surrogate_text(tmp_path, monkeypatch):
+ monkeypatch.setattr(
+ compactor,
+ "get_chat_folder_path",
+ lambda _ctxid: str(tmp_path),
+ )
+ monkeypatch.setattr(
+ compactor,
+ "export_json_chat",
+ lambda _context: '{"content":"before\ud83dafter"}',
+ )
+
+ paths = compactor._save_pre_compaction_backup(
+ SimpleNamespace(id="surrogate-chat"),
+ "transcript before\ud83dafter",
+ )
+
+ json_backup = Path(paths["json"]).read_text(encoding="utf-8")
+ text_backup = Path(paths["txt"]).read_text(encoding="utf-8")
+
+ assert "\ud83d" not in json_backup
+ assert "\ud83d" not in text_backup
+ assert "before?after" in json_backup
+ assert "before?after" in text_backup
+
+
def test_compaction_splitter_wraps_single_line_85k_payload(monkeypatch):
monkeypatch.setattr(
compactor.tokens, "approximate_tokens", lambda text: len(text or "")