Surface Generate errors in the guide panel; raise LLM max_tokens to 4096

Generate failures (missing token, LLM/provider error, no steps) now render in the
guide panel itself, not only the easy-to-miss bottom status line — the likely reason
the deployed Generate looked broken (the warning was off-screen). Also raise the
default output cap so longer guides aren't truncated.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

app.py

@@ -305,15 +305,19 @@ def _generate_draft(tr, token: str, progress):
 
 
 def on_generate(transcript_text: str, transcript_obj, hf_token: str, progress=gr.Progress()):
+    # Show any failure reason in the guide panel itself (not just the bottom
+    # status line, which is easy to miss far down the page).
     token = config.apply_token(hf_token)
     if not token:
-        return "", None, "⚠️ Enter your HuggingFace token above to generate the guide."
+        m = "⚠️ Enter your HuggingFace token in the 🔑 box at the **top** of the page, then click Generate again."
+        return m, None, m
     tr = _parse_timestamped_text(transcript_text) if transcript_text.strip() else transcript_obj
     if tr is None or not tr.segments:
-        return "", None, "Transcribe the audio first (or paste a transcript)."
+        m = "⚠️ Transcribe the audio first (step 2), or paste a transcript, then generate."
+        return m, None, m
     draft, msg = _generate_draft(tr, token, progress)
     if draft is None:
-        return "", None, msg
+        return f"⚠️ {msg}", None, msg
     return _draft_to_md(draft), draft, msg + " Review, then build the DOCX."
 
 

src/config.py

@@ -103,7 +103,7 @@ def resolve_hf_token() -> str | None:
 # --- Text LLM (HF Inference API) --------------------------------------------
 LLM_MODEL = os.getenv("DOCUMAKER_LLM_MODEL", "Qwen/Qwen2.5-7B-Instruct")
 LLM_PROVIDER = os.getenv("DOCUMAKER_LLM_PROVIDER", "").strip() or None
-LLM_MAX_TOKENS = int(os.getenv("DOCUMAKER_LLM_MAX_TOKENS", "2048"))
+LLM_MAX_TOKENS = int(os.getenv("DOCUMAKER_LLM_MAX_TOKENS", "4096"))
 LLM_TEMPERATURE = float(os.getenv("DOCUMAKER_LLM_TEMPERATURE", "0.3"))
 # Approx. characters of transcript per LLM chunk (keeps prompts within context).
 LLM_CHUNK_CHARS = int(os.getenv("DOCUMAKER_LLM_CHUNK_CHARS", "6000"))