fix(skills): preserve official optional-skill frontmatter names

agent/auxiliary_client.py

@@ -1478,7 +1478,10 @@ def _try_openrouter(explicit_api_key: str = None) -> Tuple[Optional[OpenAI], Opt
     if pool_present:
         or_key = explicit_api_key or _pool_runtime_api_key(entry)
         if not or_key:
-            _mark_provider_unhealthy("openrouter", ttl=60)
+            # Missing credentials are a static availability/config condition, not
+            # a transient provider-health issue. Do NOT poison the unhealthy
+            # cache here — doing so makes later calls skip OpenRouter even after
+            # a key appears in env/pool during the same process.
             return None, None
         base_url = _pool_runtime_base_url(entry, OPENROUTER_BASE_URL) or OPENROUTER_BASE_URL
         logger.debug("Auxiliary client: OpenRouter via pool")
@@ -1487,7 +1490,7 @@ def _try_openrouter(explicit_api_key: str = None) -> Tuple[Optional[OpenAI], Opt
 
     or_key = explicit_api_key or os.getenv("OPENROUTER_API_KEY")
     if not or_key:
-        _mark_provider_unhealthy("openrouter", ttl=60)
+        # No key configured yet: fall through quietly so other providers may win.
         return None, None
     logger.debug("Auxiliary client: OpenRouter")
     return OpenAI(api_key=or_key, base_url=OPENROUTER_BASE_URL,
@@ -1527,11 +1530,8 @@ def _try_nous(vision: bool = False) -> Tuple[Optional[OpenAI], Optional[str]]:
     nous = _read_nous_auth()
     runtime = _resolve_nous_runtime_api(force_refresh=False)
     if runtime is None and not nous:
-        logger.warning(
-            "Auxiliary Nous client unavailable: no Nous authentication found "
-            "(run: hermes auth)."
-        )
-        _mark_provider_unhealthy("nous", ttl=60)
+        # No Nous auth/runtime configured: this is simple unavailability, not a
+        # health failure. Let the chain fall through without poisoning cache.
         return None, None
     if runtime is None and nous:
         # Runtime credential mint failed but stored Nous auth is still present.
@@ -4123,6 +4123,7 @@ def _build_call_kwargs(
         "model": model,
         "messages": messages,
         "timeout": timeout,
+        "stream": False,
     }
 
     fixed_temperature = _fixed_temperature_for_model(model, base_url)

gateway/platforms/api_server.py

@@ -3115,6 +3115,47 @@ def _run_sync():
         response_headers = (
             {"X-Hermes-Session-Key": gateway_session_key} if gateway_session_key else {}
         )
+        accept_header = request.headers.get("Accept", "").lower()
+        if body.get("stream") is True and "text/event-stream" in accept_header:
+            response = web.StreamResponse(
+                status=200,
+                headers={
+                    **response_headers,
+                    "Content-Type": "text/event-stream",
+                    "Cache-Control": "no-cache",
+                    "X-Accel-Buffering": "no",
+                },
+            )
+            await response.prepare(request)
+            try:
+                await response.write((
+                    "data: "
+                    + json.dumps({
+                        "event": "response.created",
+                        "run_id": run_id,
+                        "session_id": session_id,
+                        "timestamp": created_at,
+                    })
+                    + "\n\n"
+                ).encode())
+                while True:
+                    try:
+                        event = await asyncio.wait_for(q.get(), timeout=30.0)
+                    except asyncio.TimeoutError:
+                        await response.write(b": keepalive\n\n")
+                        continue
+                    if event is None:
+                        await response.write(b": stream closed\n\n")
+                        break
+                    payload = f"data: {json.dumps(event)}\n\n"
+                    await response.write(payload.encode())
+            except Exception as exc:
+                logger.debug("[api_server] inline SSE stream error for run %s: %s", run_id, exc)
+            finally:
+                self._run_streams.pop(run_id, None)
+                self._run_streams_created.pop(run_id, None)
+            return response
+
         return web.json_response(
             {"run_id": run_id, "status": "started"},
             status=202,