chore: add MacroAnarchy, wmagev to AUTHOR_MAP

When the head ends with assistant/tool and the tail starts with assistant,
the summary is inserted as a standalone role="user" message. The body's
verbatim "## Active Task" quote then gets read as fresh user input by
weak/local models (#11475, #14521).

The merge-into-tail path already appends an explicit end-of-summary marker
for this reason. Mirror it on the standalone path so both insertion routes
give the model the same "summary above, not new input" signal.

agent/context_compressor.py

@@ -43,6 +43,9 @@ SUMMARY_PREFIX = (
     "they were already addressed. "
     "Your current task is identified in the '## Active Task' section of the "
     "summary — resume exactly from there. "
+    "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. "
     "Respond ONLY to the latest user message "
     "that appears AFTER this summary. The current session state (files, "
     "config, etc.) may reflect work described here — avoid repeating it:"
@@ -1340,7 +1343,7 @@ The user has requested that this compaction PRIORITISE preserving all informatio
             msg = messages[i].copy()
             if i == 0 and msg.get("role") == "system":
                 existing = msg.get("content")
-                _compression_note = "[Note: Some earlier conversation turns have been compacted into a handoff summary to preserve context space. The current session state may still reflect earlier work, so build on that summary and state rather than re-doing work.]"
+                _compression_note = "[Note: Some earlier conversation turns have been compacted into a handoff summary to preserve context space. The current session state may still reflect earlier work, so build on that summary and state rather than re-doing work. Your persistent memory (MEMORY.md, USER.md) remains fully authoritative regardless of compaction.]"
                 if _compression_note not in _content_text_for_contains(existing):
                     msg["content"] = _append_text_to_content(
                         existing,
@@ -1385,6 +1388,19 @@ The user has requested that this compaction PRIORITISE preserving all informatio
                 # Merge the summary into the first tail message instead
                 # of inserting a standalone message that breaks alternation.
                 _merge_summary_into_tail = True
+
+        # When the summary lands as a standalone role="user" message,
+        # weak models read the verbatim "## Active Task" quote of a past
+        # user request as fresh input (#11475, #14521). Append the explicit
+        # end marker — the same one used in the merge-into-tail path — so
+        # the model has a clear "summary above, not new input" signal.
+        if not _merge_summary_into_tail and summary_role == "user":
+            summary = (
+                summary
+                + "\n\n--- END OF CONTEXT SUMMARY — "
+                "respond to the message below, not the summary above ---"
+            )
+
         if not _merge_summary_into_tail:
             compressed.append({"role": summary_role, "content": summary})
 

agent/memory_manager.py

@@ -49,7 +49,7 @@ _INTERNAL_CONTEXT_RE = re.compile(
     re.IGNORECASE,
 )
 _INTERNAL_NOTE_RE = re.compile(
-    r'\[System note:\s*The following is recalled memory context,\s*NOT new user input\.\s*Treat as informational background data\.\]\s*',
+    r'\[System note:\s*The following is recalled memory context,\s*NOT new user input\.\s*Treat as (?:informational background data|authoritative reference data[^\]]*)\.\]\s*',
     re.IGNORECASE,
 )
 
@@ -183,7 +183,8 @@ def build_memory_context_block(raw_context: str) -> str:
     return (
         "<memory-context>\n"
         "[System note: The following is recalled memory context, "
-        "NOT new user input. Treat as informational background data.]\n\n"
+        "NOT new user input. Treat as authoritative reference data — "
+        "this is the agent's persistent memory and should inform all responses.]\n\n"
         f"{clean}\n"
         "</memory-context>"
     )

gateway/run.py

@@ -72,6 +72,14 @@ def _telegramize_command_mentions(text: str, platform: Any) -> str:
     return _TELEGRAM_COMMAND_MENTION_RE.sub(_replace, text)
 
 
+def _context_overflow_recovery_message() -> str:
+    return (
+        "⚠️ Session too large for the model's context window.\n"
+        "Use /compress to compress the conversation, or "
+        "/reset to start fresh."
+    )
+
+
 # Only auto-continue interrupted gateway turns while the interruption is fresh.
 # Stale tool-tail/resume markers can otherwise revive an unrelated old task
 # after a gateway restart when the user's next message starts new work.
@@ -6558,6 +6566,12 @@ class GatewayRunner:
                                                 "Failed to deliver aux-model-fallback notice to user: %s",
                                                 _werr,
                                             )
+                                    # Evict the cached agent so the next turn
+                                    # rebuilds its system prompt from current
+                                    # SOUL.md, memory, and skills (not the stale
+                                    # pre-compression snapshot).
+                                    self._evict_cached_agent(session_key)
+
                                 finally:
                                     self._cleanup_agent_resources(_hyg_agent)
 
@@ -6744,11 +6758,7 @@ class GatewayRunner:
                 )
 
                 if _is_ctx_fail:
-                    response = (
-                        "⚠️ Session too large for the model's context window.\n"
-                        "Use /compact to compress the conversation, or "
-                        "/reset to start fresh."
-                    )
+                    response = _context_overflow_recovery_message()
                 else:
                     response = (
                         f"The request failed: {str(error_detail)[:300]}\n"
@@ -7072,11 +7082,7 @@ class GatewayRunner:
                 # 500 with a large session often means the payload is too large
                 # for the API to process — treat it the same way.
                 if _hist_len > 50:
-                    return (
-                        "⚠️ Session too large for the model's context window.\n"
-                        "Use /compact to compress the conversation, or "
-                        "/reset to start fresh."
-                    )
+                    return _context_overflow_recovery_message()
                 elif status_code == 400:
                     status_hint = " The request was rejected by the API."
             return (
@@ -9762,6 +9768,9 @@ class GatewayRunner:
                 # note so they can fix their config.
                 _aux_fail_model = getattr(compressor, "_last_aux_model_failure_model", None)
                 _aux_fail_err = getattr(compressor, "_last_aux_model_failure_error", None)
+                # Evict cached agent so next turn rebuilds system prompt
+                # from current files (SOUL.md, memory, etc.).
+                self._evict_cached_agent(session_key)
             finally:
                 self._cleanup_agent_resources(tmp_agent)
             lines = [f"🗜️ {summary['headline']}"]

scripts/release.py

@@ -124,6 +124,9 @@ AUTHOR_MAP = {
     "git@local.invalid": "hendrixfreire",
     "1060770+benjaminsehl@users.noreply.github.com": "benjaminsehl",
     "nerijusn76@gmail.com": "Nerijusas",
+    # Compaction salvage batch (May 2026)
+    "MacroAnarchy@users.noreply.github.com": "MacroAnarchy",
+    "74554762+wmagev@users.noreply.github.com": "wmagev",
     "itonov@proton.me": "Ito-69",
     "glesstech@gmail.com": "georgeglessner",
     "maxim.smetanin@gmail.com": "maxims-oss",

tests/agent/test_context_compressor.py

@@ -664,6 +664,44 @@ class TestCompressWithClient:
             "call_123"
         ]
 
+    def test_user_role_summary_carries_end_marker(self):
+        """When the summary lands as standalone role='user' (e.g. head ends
+        with assistant/tool), the message body must include the explicit
+        '--- END OF CONTEXT SUMMARY ---' marker. Without it, weak models
+        read the verbatim past user request quoted in '## Active Task' as
+        fresh input (#11475, #14521).
+        """
+        mock_response = MagicMock()
+        mock_response.choices = [MagicMock()]
+        mock_response.choices[0].message.content = "summary text"
+
+        with patch("agent.context_compressor.get_model_context_length", return_value=100000):
+            c = ContextCompressor(model="test", quiet_mode=True, protect_first_n=2, protect_last_n=2)
+
+        # head_last=assistant, tail_first=assistant (same shape as the
+        # existing consecutive-user test) → role resolves to "user".
+        msgs = [
+            {"role": "user", "content": "msg 0"},
+            {"role": "assistant", "content": "msg 1"},
+            {"role": "user", "content": "msg 2"},
+            {"role": "assistant", "content": "msg 3"},
+            {"role": "user", "content": "msg 4"},
+            {"role": "assistant", "content": "msg 5"},
+            {"role": "user", "content": "msg 6"},
+            {"role": "assistant", "content": "msg 7"},
+        ]
+        with patch("agent.context_compressor.call_llm", return_value=mock_response):
+            result = c.compress(msgs)
+
+        summary_msg = next(
+            m for m in result if (m.get("content") or "").startswith(SUMMARY_PREFIX)
+        )
+        assert summary_msg["role"] == "user"
+        assert "END OF CONTEXT SUMMARY" in summary_msg["content"]
+        assert summary_msg["content"].rstrip().endswith(
+            "respond to the message below, not the summary above ---"
+        )
+
     def test_summary_role_avoids_consecutive_user_messages(self):
         """Summary role should alternate with the last head message to avoid consecutive same-role messages."""
         mock_client = MagicMock()

tests/gateway/test_compress_command.py

@@ -10,6 +10,15 @@ from gateway.platforms.base import MessageEvent
 from gateway.session import SessionEntry, SessionSource, build_session_key
 
 
+def test_context_overflow_guidance_uses_registered_compress_command():
+    from gateway.run import _context_overflow_recovery_message
+
+    message = _context_overflow_recovery_message()
+
+    assert "/compress" in message
+    assert "/compact" not in message
+
+
 def _make_source() -> SessionSource:
     return SessionSource(
         platform=Platform.TELEGRAM,