diff --git a/lib/crewai/src/crewai/utilities/reasoning_handler.py b/lib/crewai/src/crewai/utilities/reasoning_handler.py
index 1028a3f3de..5ddef4aad1 100644
--- a/lib/crewai/src/crewai/utilities/reasoning_handler.py
+++ b/lib/crewai/src/crewai/utilities/reasoning_handler.py
@@ -4,6 +4,7 @@
 
 import json
 import logging
+import re
 from typing import TYPE_CHECKING, Any, Final, Literal, cast
 
 from pydantic import BaseModel, Field
@@ -27,6 +28,36 @@
     from crewai.task import Task
 
 
+_READY_MARKER: Final[re.Pattern[str]] = re.compile(
+    r"\bNOT\s+READY\b|\bREADY\b", re.IGNORECASE
+)
+
+
+def _detect_plan_ready(text: str) -> bool:
+    """Detect whether a text reasoning plan concludes with READY.
+
+    The reasoning prompts instruct the model to conclude with "READY" or
+    "NOT READY", but models (especially on the text-parsing path) often emit a
+    bare ``READY`` rather than the full "READY: I am ready to execute the task."
+    sentence, which the previous exact-substring check missed (#6204).
+
+    Matches the LAST readiness marker so a "not ready" mentioned earlier in the
+    plan doesn't override a final "READY", and so "NOT READY" wins over the bare
+    "READY" it contains.
+
+    Args:
+        text: The raw model response.
+
+    Returns:
+        True if the concluding marker is READY, False for NOT READY or if no
+        marker is present.
+    """
+    markers = _READY_MARKER.findall(text)
+    if not markers:
+        return False
+    return "NOT" not in markers[-1].upper()
+
+
 class ReasoningPlan(BaseModel):
     """Model representing a reasoning plan for a task."""
 
@@ -409,7 +440,7 @@ def _create_reasoning_plan(
             return (
                 response_str,
                 [],
-                "READY: I am ready to execute the task." in response_str,
+                _detect_plan_ready(response_str),
             )
 
         except Exception as e:
@@ -433,7 +464,7 @@ def _create_reasoning_plan(
                 return (
                     fallback_str,
                     [],
-                    "READY: I am ready to execute the task." in fallback_str,
+                    _detect_plan_ready(fallback_str),
                 )
             except Exception as inner_e:
                 self.logger.error(f"Error during fallback text parsing: {inner_e!s}")
@@ -593,7 +624,7 @@ def _parse_planning_response(response: str) -> tuple[str, bool]:
             return "No plan was generated.", False
 
         plan = response
-        ready = "READY: I am ready to execute the task." in response
+        ready = _detect_plan_ready(response)
 
         return plan, ready
 
diff --git a/lib/crewai/tests/utilities/test_structured_planning.py b/lib/crewai/tests/utilities/test_structured_planning.py
index 1e36c6de97..8f03737b0c 100644
--- a/lib/crewai/tests/utilities/test_structured_planning.py
+++ b/lib/crewai/tests/utilities/test_structured_planning.py
@@ -17,9 +17,52 @@
     FUNCTION_SCHEMA,
     AgentReasoning,
     ReasoningPlan,
+    _detect_plan_ready,
 )
 
 
+class TestDetectPlanReady:
+    """Readiness detection must accept a bare READY marker, not only the full
+    'READY: I am ready to execute the task.' sentence (#6204)."""
+
+    def test_bare_ready_marker_is_ready(self):
+        assert _detect_plan_ready("1. do X\n2. do Y\n\n---\n\nREADY") is True
+
+    def test_full_sentence_still_ready(self):
+        assert _detect_plan_ready("plan...\nREADY: I am ready to execute the task.")
+
+    def test_not_ready_marker_is_not_ready(self):
+        assert (
+            _detect_plan_ready("plan...\n\nNOT READY: I need more detail.") is False
+        )
+
+    def test_last_marker_wins_over_mid_text_not_ready(self):
+        text = "Step 1 is not ready to run yet without input.\n\nREADY"
+        assert _detect_plan_ready(text) is True
+
+    def test_case_insensitive(self):
+        assert _detect_plan_ready("plan...\n\nready") is True
+        assert _detect_plan_ready("plan...\n\nnot ready") is False
+
+    def test_no_marker_defaults_to_not_ready(self):
+        assert _detect_plan_ready("just a plan with no verdict") is False
+
+    def test_ready_substring_inside_words_is_ignored(self):
+        # "already" contains "ready" but is not a READY marker (word boundaries).
+        assert _detect_plan_ready("I am already prepared") is False
+        assert _detect_plan_ready("The steps are already documented") is False
+
+    def test_trailing_already_does_not_flip_not_ready(self):
+        # A trailing "already" must not override a concluding NOT READY.
+        assert (
+            _detect_plan_ready("NOT READY. I already have the tools.") is False
+        )
+
+    def test_parse_planning_response_detects_bare_ready(self):
+        _plan, ready = AgentReasoning._parse_planning_response("my plan\n\nREADY")
+        assert ready is True
+
+
 class TestFunctionSchema:
     """Tests for the FUNCTION_SCHEMA used in structured planning."""