cat-stack 2.2.0__tar.gz → 2.3.0__tar.gz

{cat_stack-2.2.0 → cat_stack-2.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cat-stack
-Version: 2.2.0
+Version: 2.3.0
 Summary: Domain-agnostic text, image, PDF, and DOCX classification engine powered by LLMs
 Project-URL: Documentation, https://github.com/chrissoria/cat-stack#readme
 Project-URL: Issues, https://github.com/chrissoria/cat-stack/issues
@@ -23,7 +23,7 @@ Requires-Dist: pandas
 Requires-Dist: requests
 Requires-Dist: tqdm
 Provides-Extra: agent
-Requires-Dist: cat-claws>=0.1.0; extra == 'agent'
+Requires-Dist: cat-claws>=0.2.0; extra == 'agent'
 Provides-Extra: docx
 Requires-Dist: python-docx>=1.0.0; extra == 'docx'
 Provides-Extra: embeddings

{cat_stack-2.2.0 → cat_stack-2.3.0}/pyproject.toml

@@ -35,7 +35,7 @@ pdf = ["PyMuPDF>=1.23.0"]
 docx = ["python-docx>=1.0.0"]
 formatter = ["torch>=2.0.0", "transformers>=4.40.0", "accelerate>=0.27.0"]
 embeddings = ["sentence-transformers>=2.2.0"]
-agent = ["cat-claws>=0.1.0"]
+agent = ["cat-claws>=0.2.0"]
 
 [project.urls]
 Documentation = "https://github.com/chrissoria/cat-stack#readme"

{cat_stack-2.2.0 → cat_stack-2.3.0}/src/catstack/__about__.py

@@ -1,7 +1,7 @@
 # SPDX-FileCopyrightText: 2025-present Christopher Soria <chrissoria@berkeley.edu>
 #
 # SPDX-License-Identifier: GPL-3.0-or-later
-__version__ = "2.2.0"
+__version__ = "2.3.0"
 __author__ = "Chris Soria"
 __email__ = "chrissoria@berkeley.edu"
 __title__ = "cat-stack"

{cat_stack-2.2.0 → cat_stack-2.3.0}/src/catstack/image_functions.py

@@ -148,7 +148,12 @@ def image_multi_class(
         raise FileNotFoundError(f"Directory {save_directory} doesn't exist")
 
     model_source = _detect_model_source(user_model, model_source)
-    _require_http_provider(model_source, "Image classification")
+    if model_source == "claude-code":
+        raise ValueError(
+            "Image classification is not supported with model_source='claude-code' "
+            "(the text-only CLI shim). Use model_source='claude-agent' (the cat-claws "
+            "subscription backend) or an API-key provider."
+        )
 
     image_files = _load_image_files(image_input)
 
@@ -663,6 +668,27 @@ Provide the final categorization in the same JSON format:"""
 
         return """{"1":"e"}""", "Max retries exceeded"
 
+    def _call_claude_agent_image(base_text, encoded, media_type):
+        """Image classification via the cat-claws multimodal adapter (Claude
+        subscription, no API key). Returns (reply, error) like _call_anthropic."""
+        try:
+            from catclaws._adapters import get_adapter
+        except ImportError:
+            return None, ("cat-claws is not installed. Run: pip install cat-stack[agent]")
+        import asyncio
+        adapter = get_adapter("claude")
+        _system = ("You are an image classification engine. Follow the user's "
+                   "instructions exactly and reply with only what they ask for.")
+        try:
+            reply, error = asyncio.run(adapter.one_shot(
+                base_text, system_prompt=_system, model=user_model,
+                thinking_budget=thinking_budget or 0,
+                images=[{"media_type": media_type, "data": encoded}],
+            ))
+            return (None, error) if error else (reply, None)
+        except Exception as e:
+            return None, f"cat-claws image call failed: {e}"
+
     def _process_single_image(img_path):
         """Process a single image and return (reply, error_msg)."""
         encoded, ext, is_valid = _encode_image(img_path)
@@ -703,6 +729,10 @@ Provide the final categorization in the same JSON format:"""
             image_content = {"type": "image_url", "image_url": {"url": encoded_image, "detail": "high"}}
             return _call_mistral(prompt, step2_prompt, step3_prompt, step4_prompt, image_content)
 
+        elif model_source == "claude-agent":
+            media_type = f"image/{ext}" if ext else "image/jpeg"
+            return _call_claude_agent_image(base_prompt_text, encoded, media_type)
+
         else:
             raise ValueError("Unknown source! Choose from OpenAI, Anthropic, Perplexity, Google, xAI, Huggingface, or Mistral")
 

{cat_stack-2.2.0 → cat_stack-2.3.0}/src/catstack/pdf_functions.py

@@ -393,7 +393,12 @@ def pdf_multi_class(
         raise ValueError(f"mode must be 'image', 'text', or 'both', got: {mode}")
 
     model_source = _detect_model_source(user_model, model_source)
-    _require_http_provider(model_source, "PDF classification")
+    if model_source == "claude-code":
+        raise ValueError(
+            "PDF classification is not supported with model_source='claude-code' "
+            "(the text-only CLI shim). Use model_source='claude-agent' (the cat-claws "
+            "subscription backend) or an API-key provider."
+        )
 
     # Providers with native PDF support (only used in image/both modes)
     native_pdf_providers = {"anthropic", "google"}
@@ -1181,6 +1186,28 @@ Provide the final categorization in the same JSON format:"""
 
         return """{"1":"e"}""", "Max retries exceeded"
 
+    def _call_claude_agent_pdf(base_text, encoded_image):
+        """PDF-page classification via the cat-claws multimodal adapter. The page
+        is rendered to an image (PDF-as-images); Claude subscription, no API key.
+        Returns (reply, error)."""
+        try:
+            from catclaws._adapters import get_adapter
+        except ImportError:
+            return None, ("cat-claws is not installed. Run: pip install cat-stack[agent]")
+        import asyncio
+        adapter = get_adapter("claude")
+        _system = ("You are a document-page classification engine. Follow the "
+                   "user's instructions exactly and reply with only what they ask for.")
+        try:
+            reply, error = asyncio.run(adapter.one_shot(
+                base_text, system_prompt=_system, model=user_model,
+                thinking_budget=thinking_budget or 0,
+                images=[{"media_type": "image/png", "data": encoded_image}],
+            ))
+            return (None, error) if error else (reply, None)
+        except Exception as e:
+            return None, f"cat-claws PDF call failed: {e}"
+
     def _process_single_page(pdf_path, page_index, page_label):
         """Process a single PDF page and return (reply, error_msg)."""
 
@@ -1260,6 +1287,13 @@ Provide the final categorization in the same JSON format:"""
             prompt_data = _build_prompt_google_pdf(encoded_pdf, base_prompt_text)
             return _call_google(prompt_data, step2_prompt, step3_prompt, step4_prompt, base_prompt_text)
 
+        elif model_source == "claude-agent":
+            image_bytes, is_valid = _extract_page_as_image_bytes(pdf_path, page_index)
+            if not is_valid:
+                return None, "Failed to render PDF page to image"
+            encoded_image = _encode_bytes_to_base64(image_bytes)
+            return _call_claude_agent_pdf(base_prompt_text, encoded_image)
+
         # Handle providers requiring image conversion
         else:
             image_bytes, is_valid = _extract_page_as_image_bytes(pdf_path, page_index)