From 4d1b2dcf652d31b99f8e6fe9b0ca4a41e3393da2 Mon Sep 17 00:00:00 2001
From: kshitijk4poor <82637225+kshitijk4poor@users.noreply.github.com>
Date: Sat, 3 Oct 2026 15:45:41 +0530
Subject: [PATCH] fix(compaction): tolerate content-less user item at
 over-budget boundary

BN01 fold2 finding (2c Low): collapsing the over-budget branch in
d24e5875b7 read `item["content"]` unguarded. A user message whose text
is measured only via `output_text` (no `content` key) reaches that
branch when it exceeds the budget and raised KeyError, aborting the
prune. Use `item.get("content")` so it falls through to the
non-truncatable path (keep nothing, budget reset) as before the
refactor. The typed-boundary parametrize gains a row for it.
---
 agent/native_compaction.py                                 | 2 +-
 tests/agent/test_native_compaction_typed_user_retention.py | 7 +++++--
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/agent/native_compaction.py b/agent/native_compaction.py
index 67236ecca1..44c721d397 100644
--- a/agent/native_compaction.py
+++ b/agent/native_compaction.py
@@ -336,7 +336,7 @@ def prune_pre_checkpoint_items(
                 retained_reversed.append(item)
                 user_remaining -= cost
             else:
-                content = item["content"]
+                content = item.get("content")
                 if isinstance(content, str):
                     head = content[: user_remaining * 4]
                     keep = bool(head.strip())
diff --git a/tests/agent/test_native_compaction_typed_user_retention.py b/tests/agent/test_native_compaction_typed_user_retention.py
index 82071af960..cf300eadfe 100644
--- a/tests/agent/test_native_compaction_typed_user_retention.py
+++ b/tests/agent/test_native_compaction_typed_user_retention.py
@@ -58,13 +58,16 @@ _IMAGE = {"type": "input_image", "image_url": "data:image/png;base64,AAAA"}
           {"type": "input_text", "text": "trailing"}], 2, ("abcd", "efgh")),
         ([{"type": "input_text", "text": "你好世界"}], 2, ("你好",)),
         ([{"type": "input_text", "text": "CURRENT" * 20}, _IMAGE], 2, None),
+        (None, 2, None),
     ],
-    ids=["multipart-boundary", "cjk-estimator-budget", "oversized-mixed-image"],
+    ids=["multipart-boundary", "cjk-estimator-budget", "oversized-mixed-image", "oversized-output-text-only"],
 )
 def test_typed_boundary_is_budgeted_and_never_substitutes_older_ask(content, budget, expected):
     older = {"role": "user", "content": "old"}
     current = {"type": "message", "role": "user", "id": "current-user", "content": deepcopy(content)}
-    for index, part in enumerate(current["content"]):
+    if content is None:  # Content-less item measured only via output_text.
+        current = {"type": "message", "role": "user", "output_text": "CURRENT" * 20}
+    for index, part in enumerate(current.get("content") or []):
         part["metadata"] = {"index": index}
     items = [older, current, _CHECKPOINT]
     original = deepcopy(items)