diff --git a/src/llama_stack/providers/inline/responses/builtin/responses/streaming.py b/src/llama_stack/providers/inline/responses/builtin/responses/streaming.py
index 734e1b90ac..6b81563aea 100644
--- a/src/llama_stack/providers/inline/responses/builtin/responses/streaming.py
+++ b/src/llama_stack/providers/inline/responses/builtin/responses/streaming.py
@@ -1559,10 +1559,10 @@ def _approval_required(self, tool_name: str) -> bool:
             return True
         if mcp_server.require_approval == "never":
             return False
-        if isinstance(mcp_server, ApprovalFilter):
-            if mcp_server.always and tool_name in mcp_server.always:
+        if isinstance(mcp_server.require_approval, ApprovalFilter):
+            if mcp_server.require_approval.always and tool_name in mcp_server.require_approval.always:
                 return True
-            if mcp_server.never and tool_name in mcp_server.never:
+            if mcp_server.require_approval.never and tool_name in mcp_server.require_approval.never:
                 return False
         return True
 
diff --git a/tests/unit/providers/inline/responses/builtin/responses/test_approval_filter_bug.py b/tests/unit/providers/inline/responses/builtin/responses/test_approval_filter_bug.py
new file mode 100644
index 0000000000..1da2d3dbba
--- /dev/null
+++ b/tests/unit/providers/inline/responses/builtin/responses/test_approval_filter_bug.py
@@ -0,0 +1,112 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+"""Regression tests for ApprovalFilter in _approval_required().
+
+Verifies that the ApprovalFilter.always and ApprovalFilter.never lists
+are correctly evaluated when require_approval is an ApprovalFilter object.
+
+See: https://github.com/meta-llama/llama-stack/issues/5287
+"""
+
+from unittest.mock import AsyncMock, MagicMock
+
+from llama_stack.providers.inline.responses.builtin.responses.streaming import StreamingResponseOrchestrator
+from llama_stack.providers.inline.responses.builtin.responses.types import ChatCompletionContext, ToolContext
+from llama_stack_api.openai_responses import (
+    ApprovalFilter,
+    OpenAIResponseInputToolMCP,
+)
+
+
+def _build_orchestrator(mcp_tool_to_server: dict[str, OpenAIResponseInputToolMCP]) -> StreamingResponseOrchestrator:
+    """Build a minimal orchestrator with the given MCP tool mapping."""
+    mock_ctx = MagicMock(spec=ChatCompletionContext)
+    mock_ctx.tool_context = MagicMock(spec=ToolContext)
+    mock_ctx.tool_context.previous_tools = mcp_tool_to_server
+    mock_ctx.model = "test-model"
+    mock_ctx.messages = []
+    mock_ctx.temperature = None
+    mock_ctx.top_p = None
+    mock_ctx.frequency_penalty = None
+    mock_ctx.response_format = MagicMock()
+    mock_ctx.tool_choice = None
+
+    orchestrator = StreamingResponseOrchestrator(
+        inference_api=AsyncMock(),
+        ctx=mock_ctx,
+        response_id="resp_test",
+        created_at=0,
+        text=MagicMock(),
+        max_infer_iters=1,
+        tool_executor=MagicMock(),
+        instructions=None,
+        safety_api=None,
+    )
+    return orchestrator
+
+
+def _make_mcp_server(**kwargs) -> OpenAIResponseInputToolMCP:
+    defaults = {"server_label": "test-server", "server_url": "http://localhost:9999/mcp"}
+    defaults.update(kwargs)
+    return OpenAIResponseInputToolMCP(**defaults)
+
+
+class TestApprovalFilterCorrectness:
+    """Verify ApprovalFilter.always and .never lists are respected."""
+
+    def test_never_list_skips_approval(self):
+        mcp_server = _make_mcp_server(
+            require_approval=ApprovalFilter(never=["safe_tool"]),
+        )
+        orch = _build_orchestrator({"safe_tool": mcp_server})
+        assert orch._approval_required("safe_tool") is False
+
+    def test_always_list_requires_approval(self):
+        mcp_server = _make_mcp_server(
+            require_approval=ApprovalFilter(always=["dangerous_tool"]),
+        )
+        orch = _build_orchestrator({"dangerous_tool": mcp_server})
+        assert orch._approval_required("dangerous_tool") is True
+
+    def test_unlisted_tool_defaults_to_approval(self):
+        mcp_server = _make_mcp_server(
+            require_approval=ApprovalFilter(always=["other"], never=["other2"]),
+        )
+        orch = _build_orchestrator({"unlisted_tool": mcp_server})
+        assert orch._approval_required("unlisted_tool") is True
+
+    def test_unknown_tool_returns_false(self):
+        orch = _build_orchestrator({})
+        assert orch._approval_required("nonexistent") is False
+
+    def test_literal_always_requires_approval(self):
+        mcp_server = _make_mcp_server(require_approval="always")
+        orch = _build_orchestrator({"any_tool": mcp_server})
+        assert orch._approval_required("any_tool") is True
+
+    def test_literal_never_skips_approval(self):
+        mcp_server = _make_mcp_server(require_approval="never")
+        orch = _build_orchestrator({"any_tool": mcp_server})
+        assert orch._approval_required("any_tool") is False
+
+    def test_filter_with_both_lists(self):
+        mcp_server = _make_mcp_server(
+            require_approval=ApprovalFilter(
+                always=["dangerous_tool"],
+                never=["safe_tool"],
+            ),
+        )
+        mapping = {
+            "safe_tool": mcp_server,
+            "dangerous_tool": mcp_server,
+            "unlisted_tool": mcp_server,
+        }
+        orch = _build_orchestrator(mapping)
+
+        assert orch._approval_required("safe_tool") is False
+        assert orch._approval_required("dangerous_tool") is True
+        assert orch._approval_required("unlisted_tool") is True