From f2e62f60e68fb16f3df9a9434cbae04f8b9733bc Mon Sep 17 00:00:00 2001
From: root <root@hermes.chiabur.xyz>
Date: Tue, 16 Jun 2026 21:16:45 +0000
Subject: [PATCH] fix: /api/show GET support, /v1 root handler, and proxy debug
 logging

Three changes to debug and fix Hermes Desktop integration:

1. /api/show: Added GET handler alongside existing POST handler.
   Hermes Desktop probes with GET ?model=xxx, not POST body.
   Refactored shared lookup logic into _ollama_show_lookup().

2. /v1 root: Added handler returning basic info. Hermes Desktop
   probes this URL and ERR_CONNECTION_REFUSED was blocking
   full provider validation.

3. Proxy execute(): Added debug logging for non-200 responses.
   Prints the backend URL, status code, and first 500 bytes of body
   to help diagnose why llama-server returns 400 on
   /v1/chat/completions.
---
 main.py | 32 ++++++++++++++++++++++++++++++--
 1 file changed, 30 insertions(+), 2 deletions(-)

diff --git a/main.py b/main.py
index 1e371fc..53d2b3f 100644
--- a/main.py
+++ b/main.py
@@ -153,6 +153,12 @@ app = FastAPI(lifespan=lifespan)
 
 
 # ─── GET /v1/models — Issue #2 ──────────────────────────────────────────────
+@app.get("/v1")
+async def v1_root():
+    """OpenAI API root — return basic info for Hermes Desktop WebUI probe."""
+    return {"object": "list", "data": []}
+
+
 @app.get("/v1/models")
 async def get_models():
     """OpenAI-compatible /v1/models endpoint proxying to Sidecar."""
@@ -226,13 +232,26 @@ async def ollama_tags():
     return {"models": models}
 
 
+@app.get("/api/show")
+async def ollama_show_get(model: str = ""):
+    """Ollama-compatible model info for Hermes Desktop discovery (GET variant).
+
+    Some Hermes Desktop versions probe /api/show via GET with a ?model= parameter.
+    """
+    return await _ollama_show_lookup(model)
+
+
 @app.post("/api/show")
-async def ollama_show(request: Request):
-    """Ollama-compatible model info for Hermes Desktop discovery."""
+async def ollama_show_post(request: Request):
+    """Ollama-compatible model info for Hermes Desktop discovery (POST variant)."""
     body = await request.body()
     body_data = json.loads(body) if body else {}
     model_name = body_data.get("model", "")
+    return await _ollama_show_lookup(model_name)
 
+
+async def _ollama_show_lookup(model_name: str):
+    """Shared logic for Ollama /api/show model info lookup."""
     async with httpx.AsyncClient(timeout=5.0) as client:
         try:
             resp = await client.get(f"{SIDECAR_URL}/models/available")
@@ -507,8 +526,11 @@ async def proxy(
                         request.method, target,
                         content=body, headers=headers,
                     ) as resp:
+                        if resp.status_code != 200:
+                            print(f"PROXY: {target} returned {resp.status_code} during SSE stream", flush=True)
                         async for chunk in resp.aiter_bytes():
                             yield chunk
+
                 return StreamingResponse(gen(), status_code=200)
 
             resp = await client.request(
@@ -517,6 +539,12 @@ async def proxy(
                 content=body,
                 headers=headers,
             )
+            if resp.status_code != 200:
+                body_preview = resp.content[:500].decode("utf-8", errors="replace")
+                print(
+                    f"PROXY: {request.method} {target} returned {resp.status_code}: {body_preview}",
+                    flush=True,
+                )
             return Response(
                 content=resp.content,
                 status_code=resp.status_code,