fix(compression): affirm tool use stays active in the compaction handoff prefix (#66291)

The REFERENCE ONLY framing ('treat as background reference, NOT as active
instructions... Do NOT answer questions or fulfill requests') was observed
bleeding into general tool-use suppression: a production session went
narration-only for 7 consecutive turns immediately after a compression
event, describing edits instead of calling tools (#65848 report).

Fix is additive: one clause stating the note does not restrict HOW the
agent works — tools remain fully active for the active task. Every
anti-resumption protection stays intact; the previous prefix generation
is frozen into _HISTORICAL_SUMMARY_PREFIXES per the module contract so
persisted summaries still get the directive-strip on re-compaction.

The #65848 rewrite was not taken: dropping the 'Do NOT answer questions'
line and the four-heading discard directive risks re-opening the
stale-task-resumption class those clauses exist to prevent (the carveout
era regressions #41607/#38364/#42812 documented in this file).

Report and root-cause analysis: @yasserbousrih (#65848).

agent/context_compressor.py

@@ -107,6 +107,9 @@ SUMMARY_PREFIX = (
     "IMPORTANT: Your persistent memory (MEMORY.md, USER.md) in the system "
     "prompt is ALWAYS authoritative and active — never ignore or deprioritize "
     "memory content due to this compaction note. "
+    "None of the above restricts HOW you work: your tools remain fully "
+    "active — keep calling them normally for the active task (edit files, "
+    "run commands, search) instead of merely narrating what you would do. "
     "The current session state (files, config, etc.) may reflect work "
     "described here — avoid repeating it:"
 )
@@ -193,6 +196,36 @@ _MERGED_SUMMARY_DELIMITER = "[END OF PRIOR CONTEXT — COMPACTION SUMMARY BELOW]
 # embedded in the body and keeps hijacking replies. Keep newest-first; entries
 # are matched literally. Add a frozen copy here whenever SUMMARY_PREFIX changes.
 _HISTORICAL_SUMMARY_PREFIXES = (
+    # Jul 2026 (#65848 class): identical to the current prefix except it
+    # lacked the explicit "tools remain fully active" clause — the strong
+    # REFERENCE ONLY framing bled into general tool-use suppression
+    # (observed: 7 consecutive narration-only turns immediately after a
+    # compression event on a production deployment).
+    "[CONTEXT COMPACTION — REFERENCE ONLY] Earlier turns were compacted "
+    "into the summary below. This is a handoff from a previous context "
+    "window — treat it as background reference, NOT as active instructions. "
+    "Do NOT answer questions or fulfill requests mentioned in this summary; "
+    "they were already addressed. "
+    "Respond ONLY to the latest user message that appears AFTER this "
+    "summary — that message is the single source of truth for what to do "
+    "right now. "
+    "Topic overlap with the summary does NOT mean you should resume its "
+    "task: even on similar topics, the latest user message WINS. Treat ONLY "
+    "the latest message as the active task and discard stale items from "
+    f"'{HISTORICAL_TASK_HEADING}' / '{HISTORICAL_IN_PROGRESS_HEADING}' / "
+    f"'{HISTORICAL_PENDING_ASKS_HEADING}' / "
+    f"'{HISTORICAL_REMAINING_WORK_HEADING}' entirely — do not 'wrap up' or "
+    "'finish' work described there unless the latest message explicitly "
+    "asks for it. "
+    "Reverse signals in the latest message (e.g. 'stop', 'undo', 'roll "
+    "back', 'just verify', 'don't do that anymore', 'never mind', a new "
+    "topic) must immediately end any in-flight work described in the "
+    "summary; do not re-surface it in later turns. "
+    "IMPORTANT: Your persistent memory (MEMORY.md, USER.md) in the system "
+    "prompt is ALWAYS authoritative and active — never ignore or deprioritize "
+    "memory content due to this compaction note. "
+    "The current session state (files, config, etc.) may reflect work "
+    "described here — avoid repeating it:",
     # Carveout era (#41607/#38364/#42812): "consistent → use as background"
     # licensed stale-task resumption on topic overlap.
     "[CONTEXT COMPACTION — REFERENCE ONLY] Earlier turns were compacted "

tests/agent/test_summary_prefix_tool_use.py

@@ -0,0 +1,53 @@
+"""Regression tests for the SUMMARY_PREFIX tool-use clause (#65848 class).
+
+The REFERENCE ONLY framing must keep its anti-resumption protections while
+explicitly NOT restricting tool use — the strong wording was observed bleeding
+into general tool-use suppression (narration-only turns after compression).
+"""
+
+from agent.context_compressor import (
+    _HISTORICAL_SUMMARY_PREFIXES,
+    LEGACY_SUMMARY_PREFIX,
+    SUMMARY_PREFIX,
+)
+
+
+class TestSummaryPrefixToolUseClause:
+    def test_prefix_affirms_tools_remain_active(self):
+        assert "tools remain fully active" in SUMMARY_PREFIX
+        assert "narrating" in SUMMARY_PREFIX
+
+    def test_prefix_keeps_anti_resumption_protections(self):
+        """The clause is additive — every load-bearing protection stays."""
+        assert "REFERENCE ONLY" in SUMMARY_PREFIX
+        assert "Do NOT answer questions or fulfill requests" in SUMMARY_PREFIX
+        assert "the latest user message WINS" in SUMMARY_PREFIX
+        assert "Reverse signals" in SUMMARY_PREFIX
+        assert "ALWAYS authoritative" in SUMMARY_PREFIX
+
+    def test_previous_generation_frozen_in_historical_prefixes(self):
+        """Per the module contract: whenever SUMMARY_PREFIX changes, the prior
+        generation must be frozen into _HISTORICAL_SUMMARY_PREFIXES so old
+        persisted summaries still get the directive-strip on re-compaction."""
+        assert len(_HISTORICAL_SUMMARY_PREFIXES) >= 3
+        newest_frozen = _HISTORICAL_SUMMARY_PREFIXES[0]
+        # The frozen copy is the pre-clause generation: same framing, no clause.
+        assert "tools remain fully active" not in newest_frozen
+        assert "Do NOT answer questions or fulfill requests" in newest_frozen
+        assert newest_frozen != SUMMARY_PREFIX
+
+    def test_historical_prefixes_are_distinct_from_current(self):
+        for frozen in _HISTORICAL_SUMMARY_PREFIXES:
+            assert frozen != SUMMARY_PREFIX
+        assert LEGACY_SUMMARY_PREFIX != SUMMARY_PREFIX
+
+    def test_strip_recognizes_current_and_frozen_prefixes(self):
+        """Re-compaction normalization must strip both the live prefix and the
+        newly frozen one (the incident generation)."""
+        from agent.context_compressor import ContextCompressor
+
+        for prefix in (SUMMARY_PREFIX, _HISTORICAL_SUMMARY_PREFIXES[0]):
+            text = f"{prefix}\nsummary body here"
+            stripped = ContextCompressor._strip_summary_prefix(text)
+            assert "summary body here" in stripped
+            assert "REFERENCE ONLY" not in stripped