Workflow1111 — Automatic1111-style diffusion studio on gr.Workflow

README.md

@@ -345,33 +345,40 @@ image, params, hires = client.predict(
 Run `python apps/05_workflow1111/test_pipelines.py local` to print the full
 endpoint list with parameter names and types.
 
-### Bring your own token (API and MCP callers)
+### Bring your own token (API, CLI and MCP callers)
 
 The Space holds **no** `HF_TOKEN` secret. In the browser, the Sign-in button
-supplies your token through OAuth; API and MCP callers send it on the request
-as an `X-HF-Token` header instead. Two hooks make that work:
+supplies your token through OAuth; every other caller sends it on the request.
+Two hooks make that work:
 
 * every node that talks to Inference Providers declares a `request: gr.Request`
   parameter, and `_hf_token()` in `nodes.py` resolves the token in this order:
-  OAuth session → `X-HF-Token` header → `Authorization: Bearer` header →
-  server `HF_TOKEN` → local `hf auth login`;
+  OAuth session → `X-HF-Token` header → `X-HF-Authorization: Bearer` header →
+  `Authorization: Bearer` header → server `HF_TOKEN` → local `hf auth login`;
 * `app.py` extends gradio's own `_resolve_token` (used for `model` and `space`
   nodes, which gradio runs itself) with the same header fallback.
 
-Gradio's MCP server forwards the caller's headers verbatim, so the same header
-works for tools. Use `X-HF-Token`, not `Authorization: Bearer`: the Spaces
-proxy consumes the `Authorization` header before it reaches the app (it does
-work when running locally).
+`gradio_client` sends your token as `X-HF-Authorization` (it renames the
+`Authorization` header so the Hub proxy never sees it), so the ordinary
+`token=` argument just works — and so does the locally saved `hf auth login`
+token, which the client sends by default:
 
 ```python
 from gradio_client import Client
 
-client = Client("https://ysharma-workflow1111.hf.space",
-                headers={"X-HF-Token": "hf_..."})
+client = Client("https://ysharma-workflow1111.hf.space")          # uses `hf auth login` token
+client = Client("https://ysharma-workflow1111.hf.space", token="hf_...")
 prompt = client.predict("an orange cat with a yellow hat", api_name="/generated_prompt")
 ```
 
-MCP client config (Claude Code, Cursor, Claude Desktop):
+From the terminal (needs `typer>=0.15`; `gradio info <url>` prints the payload shapes):
+
+```bash
+gradio predict https://ysharma-workflow1111.hf.space /generated_prompt '{"in_0": "an orange cat with a yellow hat"}'
+```
+
+MCP clients forward custom headers verbatim, so use `X-HF-Token` there
+(Claude Code, Cursor, Claude Desktop):
 
 ```json
 {
@@ -384,6 +391,10 @@ MCP client config (Claude Code, Cursor, Claude Desktop):
 }
 ```
 
+Pass a non-empty *Hires refine instruction* to `/image` (for example
+`enhance fine detail`); with an empty one the Kontext hires node fails with
+`KeyError: 'images'` from the provider adapter.
+
 Without a token, model nodes raise `No Hugging Face token ...`; the purely
 local Pillow/numpy nodes never need one.
 

nodes.py

@@ -631,10 +631,12 @@ def _caller_hf_token(request):
     the caller's headers verbatim — so a client can authenticate per call
     without the Space holding a token of its own. Two spellings are accepted:
 
-      * ``X-HF-Token: hf_...``          (custom header; works everywhere)
-      * ``Authorization: Bearer hf_...`` (what ``gradio_client`` sends for
-        ``Client(..., token=...)``; fine locally, but the Spaces proxy
-        consumes this header, so on a Space use ``X-HF-Token``)
+      * ``X-HF-Token: hf_...``               (custom header, e.g. MCP configs)
+      * ``X-HF-Authorization: Bearer hf_...`` (what ``gradio_client`` and the
+        ``gradio predict`` CLI send for ``token=`` / ``--token`` / the locally
+        saved ``hf auth login`` token — the client renames ``Authorization``
+        to this so the Hub proxy never sees it)
+      * ``Authorization: Bearer hf_...``      (plain HTTP callers)
     """
     headers = getattr(request, "headers", None)
     if not headers:
@@ -642,9 +644,11 @@ def _caller_hf_token(request):
     lower = {str(k).lower(): v for k, v in dict(headers).items()}
     token = lower.get("x-hf-token")
     if not token:
-        auth = lower.get("authorization", "")
-        if auth.lower().startswith("bearer "):
-            token = auth[7:]
+        for key in ("x-hf-authorization", "authorization"):
+            auth = lower.get(key, "")
+            if auth.lower().startswith("bearer "):
+                token = auth[7:]
+                break
     token = (token or "").strip()
     return token if token.startswith("hf_") else None