feat(gemini): add X-Goog-Api-Client headers for API telemetry

Add `X-Goog-Api-Client` tracking headers and versioned `User-Agent`
strings to Gemini API requests across native inference adapter, tier
probes, and model probes.

This follows the Gemini API partner integration guidelines
(https://ai.google.dev/gemini-api/docs/partner-integration) and ensures
hermes-agent traffic is correctly attributed in Google Generative
Language API analytics.

Header format: `hermes-agent/<version>` (e.g., `hermes-agent/0.16.0`)

agent/gemini_native_adapter.py

@@ -31,6 +31,11 @@ from agent.gemini_schema import sanitize_gemini_tool_parameters
 
 logger = logging.getLogger(__name__)
 
+try:
+    from hermes_cli import __version__ as _HERMES_VERSION
+except Exception:
+    _HERMES_VERSION = "0.0.0"
+
 DEFAULT_GEMINI_BASE_URL = "https://generativelanguage.googleapis.com/v1beta"
 
 # Published max output-token ceiling shared by every current Gemini text model
@@ -98,7 +103,10 @@ def probe_gemini_tier(
                 url,
                 params={"key": key},
                 json=payload,
-                headers={"Content-Type": "application/json"},
+                headers={
+                    "Content-Type": "application/json",
+                    "X-Goog-Api-Client": f"hermes-agent/{_HERMES_VERSION}",
+                },
             )
     except Exception as exc:
         logger.debug("probe_gemini_tier: network error: %s", exc)
@@ -881,7 +889,8 @@ class GeminiNativeClient:
             "Content-Type": "application/json",
             "Accept": "application/json",
             "x-goog-api-key": self.api_key,
-            "User-Agent": "hermes-agent (gemini-native)",
+            "User-Agent": f"hermes-agent/{_HERMES_VERSION} (gemini-native)",
+            "X-Goog-Api-Client": f"hermes-agent/{_HERMES_VERSION}",
         }
         headers.update(self._default_headers)
         return headers

hermes_cli/models.py

@@ -3417,7 +3417,10 @@ def probe_api_models(
         candidates.append((alternate_base, True))
 
     tried: list[str] = []
-    headers: dict[str, str] = {"User-Agent": _HERMES_USER_AGENT}
+    headers: dict[str, str] = {
+        "User-Agent": _HERMES_USER_AGENT,
+        "X-Goog-Api-Client": f"hermes-agent/{_HERMES_VERSION}",
+    }
     if api_key and api_mode == "anthropic_messages":
         headers["x-api-key"] = api_key
         headers["anthropic-version"] = "2023-06-01"

tests/agent/test_gemini_native_adapter.py

@@ -408,3 +408,53 @@ def test_explicit_max_tokens_is_respected():
 
     req = build_gemini_request(messages=[{"role": "user", "content": "hi"}], max_tokens=4096)
     assert req["generationConfig"]["maxOutputTokens"] == 4096
+
+
+# ---------------------------------------------------------------------------
+# X-Goog-Api-Client header tests
+# ---------------------------------------------------------------------------
+
+
+def test_x_goog_api_client_header_is_set():
+    """The X-Goog-Api-Client header should be set on inference requests."""
+    from agent.gemini_native_adapter import GeminiNativeClient
+
+    client = GeminiNativeClient(api_key="fake-key", model="gemini-2.0-flash")
+    headers = client._headers()
+
+    assert "X-Goog-Api-Client" in headers, "X-Goog-Api-Client header missing"
+    assert "hermes-agent/" in headers["X-Goog-Api-Client"], (
+        "hermes-agent not found in X-Goog-Api-Client header"
+    )
+
+
+def test_x_goog_api_client_header_format():
+    """Header value should be 'hermes-agent/<version>' matching the package version."""
+    from agent.gemini_native_adapter import GeminiNativeClient, _HERMES_VERSION
+
+    client = GeminiNativeClient(api_key="fake-key", model="gemini-2.0-flash")
+    headers = client._headers()
+
+    expected = f"hermes-agent/{_HERMES_VERSION}"
+    assert headers["X-Goog-Api-Client"] == expected
+
+
+def test_user_agent_contains_version():
+    """User-Agent should include the hermes-agent version."""
+    from agent.gemini_native_adapter import GeminiNativeClient, _HERMES_VERSION
+
+    client = GeminiNativeClient(api_key="fake-key", model="gemini-2.0-flash")
+    headers = client._headers()
+
+    assert f"hermes-agent/{_HERMES_VERSION}" in headers["User-Agent"]
+
+
+def test_hermes_version_is_valid():
+    """_HERMES_VERSION should be a non-empty string."""
+    from agent.gemini_native_adapter import _HERMES_VERSION
+
+    assert isinstance(_HERMES_VERSION, str)
+    assert len(_HERMES_VERSION) > 0
+    assert _HERMES_VERSION != "0.0.0", (
+        "Version should resolve from hermes_cli.__version__, not the fallback"
+    )