fix(tui): preserve usage telemetry fields

nanobot/channels/websocket/tests/test_websocket_channel.py

@@ -3319,6 +3319,11 @@ async def test_send_turn_end_includes_latency_ms_when_present() -> None:
     channel = WebSocketChannel({"enabled": True, "allowFrom": ["*"]}, bus, gateway=_basic_handler(bus))
     mock_ws = AsyncMock()
     channel._attach(mock_ws, "chat-1")
+    usage = LLMUsage.reported(
+        input_tokens=80,
+        output_tokens=20,
+        cache_read_tokens=40,
+    ).with_timing(generation_ms=500, ttft_ms=125)
 
     await channel.send(OutboundMessage(
         channel="websocket",
@@ -3326,7 +3331,7 @@ async def test_send_turn_end_includes_latency_ms_when_present() -> None:
         content="",
         event=TurnEndEvent(
             latency_ms=1500,
-            usage=LLMUsage.reported(input_tokens=80, output_tokens=20, cache_read_tokens=40),
+            usage=usage,
             context_window_tokens=128_000,
         ),
     ))
@@ -3336,11 +3341,19 @@ async def test_send_turn_end_includes_latency_ms_when_present() -> None:
             "event": "turn_end",
             "chat_id": "chat-1",
             "latency_ms": 1500,
-            "usage": LLMUsage.reported(
+            "usage": {
-                input_tokens=80,
+                "prompt_tokens": 80,
-                output_tokens=20,
+                "completion_tokens": 20,
-                cache_read_tokens=40,
+                "total_tokens": 100,
-            ).to_turn_dict(),
+                "context_tokens": 80,
+                "cached_tokens": 40,
+                "request_count": 1,
+                "estimated_tokens": 0,
+                "generation_ms": 500,
+                "measured_completion_tokens": 20,
+                "ttft_ms": 125,
+                "timed_requests": 1,
+            },
             "context_window_tokens": 128_000,
         },
         {"event": "session_updated", "chat_id": "chat-1", "scope": "thread"},
@@ -5307,7 +5320,12 @@ async def test_handle_session_context_get_reads_detached_session() -> None:
 
     from nanobot.session import Session
 
-    usage = LLMUsage.reported(input_tokens=12, output_tokens=3, total_tokens=175)
+    usage = LLMUsage.reported(
+        input_tokens=12,
+        output_tokens=3,
+        total_tokens=175,
+        cache_read_tokens=6,
+    ).with_timing(generation_ms=300, ttft_ms=45)
     session = Session(
         key="websocket:context-route",
         messages=[{"role": "user", "content": "hello"}],
@@ -5328,7 +5346,19 @@ async def test_handle_session_context_get_reads_detached_session() -> None:
     assert response.status_code == 200
     body = json.loads(response.body.decode())
     assert body["replay_messages"] == 1
-    assert body["last_usage"] == usage.to_dict()
+    assert body["last_usage"] == {
+        "prompt_tokens": 12,
+        "completion_tokens": 3,
+        "total_tokens": 175,
+        "context_tokens": 12,
+        "cached_tokens": 6,
+        "request_count": 1,
+        "estimated_tokens": 0,
+        "generation_ms": 300,
+        "measured_completion_tokens": 3,
+        "ttft_ms": 45,
+        "timed_requests": 1,
+    }
     manager.read_session_snapshot.assert_called_once_with(session.key)
 
 

nanobot/providers/base.py

@@ -451,7 +451,7 @@ class LLMUsage:
         }
 
     def to_turn_dict(self) -> dict[str, int]:
-        """Project canonical usage into the WebUI's compact per-turn shape."""
+        """Project canonical usage into the compact WebUI/TUI per-turn shape."""
         result: dict[str, int] = {
             "prompt_tokens": self.input_tokens,
             "completion_tokens": self.output_tokens,
@@ -465,6 +465,12 @@ class LLMUsage:
             result["cached_tokens"] = self.cache_read_tokens
         if self.cache_write_tokens is not None:
             result["cache_write_tokens"] = self.cache_write_tokens
+        if self.generation_ms > 0 and self.measured_output_tokens > 0:
+            result["generation_ms"] = self.generation_ms
+            result["measured_completion_tokens"] = self.measured_output_tokens
+        if self.timed_requests > 0:
+            result["ttft_ms"] = self.ttft_ms
+            result["timed_requests"] = self.timed_requests
         return result
 
     @classmethod

nanobot/webui/session_context.py

@@ -38,7 +38,7 @@ def session_context_payload(session: Session) -> dict[str, Any]:
         estimate_message_tokens({"role": "system", "content": summary}) if summary else 0
     )
     stored_usage = LLMUsage.from_dict(session.metadata.get("_last_usage"))
-    last_usage = stored_usage.to_dict() if stored_usage is not None else None
+    last_usage = stored_usage.to_turn_dict() if stored_usage is not None else None
 
     return {
         "schema_version": 1,

tests/providers/test_usage_contract.py

@@ -102,7 +102,10 @@ def test_usage_projects_compact_turn_observability_shape() -> None:
         output_tokens=3,
         total_tokens=20,
         cache_read_tokens=4,
-    ) + LLMUsage.estimated(input_tokens=18, output_tokens=2)
+    ).with_timing(generation_ms=250, ttft_ms=50) + LLMUsage.estimated(
+        input_tokens=18,
+        output_tokens=2,
+    ).with_timing(generation_ms=100, ttft_ms=None)
 
     assert usage.to_turn_dict() == {
         "prompt_tokens": 30,
@@ -111,6 +114,10 @@ def test_usage_projects_compact_turn_observability_shape() -> None:
         "context_tokens": 18,
         "request_count": 2,
         "estimated_tokens": 20,
+        "generation_ms": 350,
+        "measured_completion_tokens": 5,
+        "ttft_ms": 50,
+        "timed_requests": 1,
     }
 
 

tests/webui/test_session_context.py

@@ -59,7 +59,12 @@ def test_session_context_tolerates_untrusted_summary_metadata() -> None:
 
 
 def test_session_context_sanitizes_usage_metadata() -> None:
-    usage = LLMUsage.reported(input_tokens=120, output_tokens=8, total_tokens=175)
+    usage = LLMUsage.reported(
+        input_tokens=120,
+        output_tokens=8,
+        total_tokens=175,
+        cache_read_tokens=48,
+    ).with_timing(generation_ms=400, ttft_ms=75)
     session = Session(
         key="websocket:context",
         metadata={"_last_usage": usage.to_dict()},
@@ -67,4 +72,16 @@ def test_session_context_sanitizes_usage_metadata() -> None:
 
     payload = session_context_payload(session)
 
-    assert payload["last_usage"] == usage.to_dict()
+    assert payload["last_usage"] == {
+        "prompt_tokens": 120,
+        "completion_tokens": 8,
+        "total_tokens": 175,
+        "context_tokens": 120,
+        "cached_tokens": 48,
+        "request_count": 1,
+        "estimated_tokens": 0,
+        "generation_ms": 400,
+        "measured_completion_tokens": 8,
+        "ttft_ms": 75,
+        "timed_requests": 1,
+    }

tui/src/protocol.test.ts

@@ -726,6 +726,19 @@ describe("gateway protocol", () => {
       estimated_session_tokens: 2176,
       archived_summary: "Older work was compacted.",
       archived_summary_at: "2026-08-13T10:00:00Z",
+      last_usage: {
+        prompt_tokens: 1200,
+        completion_tokens: 80,
+        total_tokens: 1280,
+        context_tokens: 1200,
+        cached_tokens: 900,
+        request_count: 1,
+        estimated_tokens: 0,
+        generation_ms: 4000,
+        measured_completion_tokens: 80,
+        ttft_ms: 250,
+        timed_requests: 1,
+      },
     })))) as unknown as typeof fetch
 
     try {
@@ -738,7 +751,19 @@ describe("gateway protocol", () => {
         estimatedSessionTokens: 2176,
         archivedSummary: "Older work was compacted.",
         archivedSummaryAt: "2026-08-13T10:00:00Z",
-        lastUsage: null,
+        lastUsage: {
+          prompt_tokens: 1200,
+          completion_tokens: 80,
+          total_tokens: 1280,
+          context_tokens: 1200,
+          cached_tokens: 900,
+          request_count: 1,
+          estimated_tokens: 0,
+          generation_ms: 4000,
+          measured_completion_tokens: 80,
+          ttft_ms: 250,
+          timed_requests: 1,
+        },
       })
     } finally {
       globalThis.fetch = original

tui/src/protocol.ts

@@ -222,7 +222,10 @@ export interface TokenUsage {
   prompt_tokens?: number
   completion_tokens?: number
   cached_tokens?: number
+  cache_write_tokens?: number
   total_tokens?: number
+  context_tokens?: number
+  request_count?: number
   provider_tokens?: number
   estimated_tokens?: number
   cost_usd?: number
@@ -372,7 +375,10 @@ function isTokenUsage(value: unknown): value is TokenUsage {
     "prompt_tokens",
     "completion_tokens",
     "cached_tokens",
+    "cache_write_tokens",
     "total_tokens",
+    "context_tokens",
+    "request_count",
     "provider_tokens",
     "estimated_tokens",
     "cost_usd",