fix(agent): repair empty-name tool_calls in sanitizer to prevent Responses 400 (salvage #12807/#52893) (#55922)

* fix(agent): drop tool_calls with empty function.name to prevent orphan 400

Salvage of #12807 by @melonboy312 — rebased onto current main (sanitizer
moved to agent_runtime_helpers), scoped to the sanitizer fix, with a
regression test that fails without it.

* fix(agent): repair (not drop) empty-name tool_calls to preserve anti-priming + prevent 400

Dropping empty-name tool_calls in the pre-call sanitizer collided with #47967,
which intentionally keeps an empty-name call paired with a synthesized
'tool name was empty' anti-priming result so weak models self-correct without
a full catalog dump. Dropping the call orphaned that result and stripped the
signal (breaking tests/agent/test_empty_tool_name_loop_dampening.py).

The actual HTTP 400 cause is an ORPHANED function_call_output (adapter drops
the empty-name function_call but keeps its output). Rename the blank name to a
non-empty sentinel instead: the call and its result stay paired, the adapter
no longer drops the function_call, no orphan, no 400 — and the anti-priming
result content the model needs is preserved.

---------

Co-authored-by: Bartok9 <danielrpike9@gmail.com>

agent/agent_runtime_helpers.py

@@ -2257,6 +2257,54 @@ def sanitize_api_messages(messages: List[Dict[str, Any]]) -> List[Dict[str, Any]
         filtered.append(msg)
     messages = filtered
 
+    # --- Repair tool_calls whose function.name is empty/missing ---
+    # Some providers (and partially-streamed responses) emit a tool_call with
+    # id="call_xxx" but function.name="". Downstream Responses-API adapters
+    # silently DROP such function_call items while still emitting the matching
+    # function_call_output, producing the gateway's HTTP 400
+    # "No tool call found for function call output with call_id ...".
+    #
+    # We do NOT drop the call: hermes' own dispatch loop intentionally keeps an
+    # empty-name call paired with a synthesized anti-priming tool result
+    # ("tool name was empty", see #47967) so weak models self-correct instead of
+    # being fed the full tool catalog. Dropping the call here would (a) orphan
+    # that result and strip the anti-priming signal, and (b) still leave any
+    # provider-side orphan. Instead, rename the blank name to a non-empty
+    # sentinel so the call and its result stay PAIRED — the adapter no longer
+    # drops the function_call, so there is no orphaned output and no 400, while
+    # the result content the model needs is preserved.
+    _EMPTY_NAME_SENTINEL = "invalid_tool_call"
+    for msg in messages:
+        if msg.get("role") != "assistant":
+            continue
+        tcs = msg.get("tool_calls") or []
+        if not tcs:
+            continue
+        for tc in tcs:
+            if isinstance(tc, dict):
+                fn = tc.get("function")
+                name = fn.get("name") if isinstance(fn, dict) else getattr(fn, "name", None)
+            else:
+                fn = getattr(tc, "function", None)
+                name = getattr(fn, "name", None) if fn else None
+            if isinstance(name, str) and name.strip():
+                continue
+            _ra().logger.warning(
+                "Pre-call sanitizer: repairing tool_call with empty "
+                "function.name -> %r (id=%s)",
+                _EMPTY_NAME_SENTINEL,
+                _ra().AIAgent._get_tool_call_id_static(tc),
+            )
+            if isinstance(fn, dict):
+                fn["name"] = _EMPTY_NAME_SENTINEL
+            elif fn is not None and hasattr(fn, "name"):
+                try:
+                    fn.name = _EMPTY_NAME_SENTINEL
+                except Exception:
+                    pass
+            elif isinstance(tc, dict):
+                tc["function"] = {"name": _EMPTY_NAME_SENTINEL, "arguments": "{}"}
+
     surviving_call_ids: set = set()
     for msg in messages:
         if msg.get("role") == "assistant":

tests/run_agent/test_run_agent.py

@@ -3659,6 +3659,48 @@ class TestHandleMaxIterations:
             "call_123"
         ]
 
+    def test_api_sanitizer_repairs_tool_call_with_empty_function_name(self, agent):
+        """A tool_call with id but empty function.name makes the Responses-API
+        adapter drop the function_call while keeping its function_call_output,
+        causing the gateway's HTTP 400 'No tool call found for function call
+        output ...'. The sanitizer renames the blank name to a non-empty
+        sentinel so the call and its result stay PAIRED (no orphaned output,
+        no 400) while the result content is preserved — it must NOT drop the
+        call, because hermes' dispatch loop keeps empty-name calls paired with
+        an anti-priming result for self-correction (#47967). (#12807)"""
+        messages = [
+            {
+                "role": "assistant",
+                "content": "",
+                "tool_calls": [
+                    {
+                        "id": "call_good",
+                        "type": "function",
+                        "function": {"name": "web_search", "arguments": "{}"},
+                    },
+                    {
+                        "id": "call_bad",
+                        "type": "function",
+                        "function": {"name": "", "arguments": "{}"},
+                    },
+                ],
+            },
+            {"role": "tool", "tool_call_id": "call_good", "content": "ok"},
+            {"role": "tool", "tool_call_id": "call_bad", "content": "orphan"},
+        ]
+
+        sanitized = agent._sanitize_api_messages(messages)
+
+        # The good call is untouched; the malformed call is repaired in place
+        # (renamed to a non-empty sentinel) rather than dropped.
+        assistant = next(m for m in sanitized if m.get("role") == "assistant")
+        names = [tc["function"]["name"] for tc in assistant["tool_calls"]]
+        assert names == ["web_search", "invalid_tool_call"]
+        # Both calls now have non-empty names, so neither output is orphaned
+        # and both tool results survive — this is what prevents the 400.
+        tool_ids = [m.get("tool_call_id") for m in sanitized if m.get("role") == "tool"]
+        assert tool_ids == ["call_good", "call_bad"]
+
 
 class TestRunConversation:
     """Tests for the main run_conversation method.