Count thoughts_fallback warning toward unusable-response limit

Context Doctor's fw.msg_thoughts_fallback.md warning was excluded from the core stop-unusable-response loop allowlist, so consecutive plain-text fallbacks never advanced the counter. Add it alongside misformat, repeat, empty_response, and reasoning_only, update the plugin DOX, and cover the behavior with a regression test.

extensions/python/_functions/agent/Agent/hist_add_warning/end/_90_stop_unusable_response_loop.py

@@ -24,6 +24,7 @@ class StopUnusableResponseLoop(Extension):
             self.agent.read_prompt("fw.msg_repeat.md"),
             self.agent.read_prompt("fw.msg_empty_response.md"),
             self.agent.read_prompt("fw.msg_reasoning_only.md"),
+            self.agent.read_prompt("fw.msg_thoughts_fallback.md"),
         }:
             return
 

plugins/_context_doctor/AGENTS.md

@@ -18,7 +18,7 @@
 - Repaired and fallback JSON is always minified.
 - Nonempty non-tool output becomes `{"thoughts":[raw]}`; XML-like output becomes `{}` only when suppression is enabled.
 - Blank-line-separated thoughts expand into separate entries after repair when the split strategy is enabled (default on).
-- A raw-text fallback conversion emits a `fw.msg_thoughts_fallback.md` history warning and a separate `fw.msg_thoughts_fallback_response.md` user notice, refreshes the generating log and response item with the transformed thoughts JSON, then sets `skip_default_processing` so the retry advances the unusable-response counter. The warning prompts are plugin-owned and are not counted by the core stop-unusable-response loop.
+- A raw-text fallback conversion emits a `fw.msg_thoughts_fallback.md` history warning and a separate `fw.msg_thoughts_fallback_response.md` user notice, refreshes the generating log and response item with the transformed thoughts JSON, then sets `skip_default_processing` so the retry advances the unusable-response counter. The fallback warning is counted by the core stop-unusable-response loop alongside `fw.msg_misformat.md`, `fw.msg_repeat.md`, `fw.msg_empty_response.md`, and `fw.msg_reasoning_only.md`.
 - Log kvps retain streamed `reasoning` and `thoughts`, then add transformed output; `update_log` controls only View Details content.
 - A repaired `response` tool call refreshes the response log item when streaming did not create it.
 - Runtime setup reads the `json_repair` pin from root `requirements.txt`; do not duplicate its version in plugin code.

tests/test_unusable_response_loop.py

@@ -101,6 +101,21 @@ def test_reasoning_only_warning_counts_toward_the_limit(monkeypatch):
     assert agent.loop_data.params_persistent[response_loop.STATE_KEY]["count"] == 1
 
 
+def test_thoughts_fallback_warning_counts_toward_the_limit(monkeypatch):
+    monkeypatch.setattr(
+        response_loop,
+        "get_settings",
+        lambda: {"max_consecutive_unusable_responses": 1},
+    )
+    agent = _agent()
+    extension = response_loop.StopUnusableResponseLoop(agent=agent)
+
+    data = _run(extension, agent, "thoughts fallback")
+
+    assert isinstance(data["exception"], HandledException)
+    assert agent.loop_data.params_persistent[response_loop.STATE_KEY]["count"] == 1
+
+
 def test_general_settings_expose_the_default_failure_limit():
     settings = get_default_settings()
     assert settings["max_consecutive_unusable_responses"] == 5