cat-stack 2.0.1__tar.gz → 2.1.1__tar.gz

{cat_stack-2.0.1 → cat_stack-2.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cat-stack
-Version: 2.0.1
+Version: 2.1.1
 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
@@ -22,6 +22,8 @@ Requires-Python: >=3.8
 Requires-Dist: pandas
 Requires-Dist: requests
 Requires-Dist: tqdm
+Provides-Extra: agent
+Requires-Dist: cat-claws>=0.1.0; extra == 'agent'
 Provides-Extra: docx
 Requires-Dist: python-docx>=1.0.0; extra == 'docx'
 Provides-Extra: embeddings

{cat_stack-2.0.1 → cat_stack-2.1.1}/pyproject.toml

@@ -35,6 +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"]
 
 [project.urls]
 Documentation = "https://github.com/chrissoria/cat-stack#readme"

{cat_stack-2.0.1 → cat_stack-2.1.1}/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__ = "2.0.1"
+__version__ = "2.1.1"
 __author__ = "Chris Soria"
 __email__ = "chrissoria@berkeley.edu"
 __title__ = "cat-stack"

{cat_stack-2.0.1 → cat_stack-2.1.1}/src/catstack/_batch.py

@@ -65,7 +65,7 @@ BATCH_ENDPOINTS = {
     },
 }
 
-UNSUPPORTED_BATCH_PROVIDERS = {"huggingface", "huggingface-together", "perplexity", "ollama"}
+UNSUPPORTED_BATCH_PROVIDERS = {"huggingface", "huggingface-together", "perplexity", "ollama", "claude-code", "claude-agent"}
 
 # Terminal states per provider
 _TERMINAL_STATES = {

{cat_stack-2.0.1 → cat_stack-2.1.1}/src/catstack/_providers.py

@@ -721,9 +721,32 @@ PROVIDER_CONFIG = {
         "auth_header": None,
         "auth_prefix": "",
     },
+    "claude-agent": {
+        "endpoint": None,  # Uses the cat-claws SDK adapter, not HTTP
+        "auth_header": None,
+        "auth_prefix": "",
+    },
 }
 
 
+# Providers that route through complete() with no HTTP endpoint of their own
+# (Claude subscription / CLI). Features that build a direct HTTP request
+# (image, PDF) can't use them — guard with a clear error, not a deep crash.
+_SUBSCRIPTION_PROVIDERS = ("claude-code", "claude-agent")
+
+
+def _require_http_provider(model_source, feature):
+    """Raise a clear error when an HTTP-only feature is used with a
+    subscription/CLI provider (claude-code / claude-agent)."""
+    if model_source in _SUBSCRIPTION_PROVIDERS:
+        raise ValueError(
+            f"{feature} is not supported with model_source='{model_source}'. "
+            "The Claude subscription/CLI backend supports text classification, "
+            "extraction, and summarization, but not " + feature.lower() + ". "
+            "Use an API-key provider (e.g. model_source='anthropic') instead."
+        )
+
+
 # =============================================================================
 # Unified API Client
 # =============================================================================
@@ -1195,7 +1218,7 @@ class UnifiedLLMClient:
                 except FileNotFoundError:
                     return None, (
                         "Claude CLI not found. Install it: "
-                        "https://docs.anthropic.com/en/docs/claude-code"
+                        "https://code.claude.com/docs"
                     )
 
             return None, "Max retries exceeded"
@@ -1206,6 +1229,55 @@ class UnifiedLLMClient:
             # contract that callers depend on.
             return None, f"Claude CLI subprocess failed: {e} (prompt may be too large for argv)"
 
+    def _call_claude_agent(
+        self,
+        messages: list,
+        thinking_budget: int = None,
+    ) -> tuple[str, str | None]:
+        """Route one completion through the cat-claws SDK adapter.
+
+        Like `_call_claude_cli`, this runs on the user's Claude subscription
+        (no API key) and returns the same (text, error) contract. cat-claws is
+        an optional dependency (the `[agent]` extra); a missing install
+        degrades to a clear install hint rather than an ImportError traceback.
+
+        The adapter is async. complete() is sync and may run inside ensemble
+        worker threads, so we drive one sealed call per invocation with
+        asyncio.run (a fresh loop per call) - never a shared/module-global
+        loop. Message flattening mirrors _call_claude_cli exactly.
+        """
+        try:
+            from catclaws._adapters import get_adapter
+        except ImportError:
+            return None, (
+                "cat-claws is not installed. Install it to use "
+                "model_source='claude-agent': pip install cat-stack[agent]"
+            )
+        import asyncio
+
+        system_parts = []
+        user_parts = []
+        for msg in messages:
+            if msg["role"] == "system":
+                system_parts.append(msg["content"])
+            elif msg["role"] in ("user", "assistant"):
+                user_parts.append(msg["content"])
+        system_prompt = "\n\n".join(system_parts) if system_parts else None
+        user_prompt = "\n\n".join(user_parts)
+
+        adapter = get_adapter("claude")
+        try:
+            return asyncio.run(
+                adapter.one_shot(
+                    user_prompt,
+                    system_prompt=system_prompt,
+                    model=self.model,
+                    thinking_budget=thinking_budget or 0,
+                )
+            )
+        except Exception as e:
+            return None, f"cat-claws call failed: {e}"
+
     def complete(
         self,
         messages: list,
@@ -1249,6 +1321,9 @@ class UnifiedLLMClient:
         if self.provider == "claude-code":
             return self._call_claude_cli(messages, max_retries=max_retries, initial_delay=initial_delay)
 
+        if self.provider == "claude-agent":
+            return self._call_claude_agent(messages, thinking_budget=thinking_budget)
+
         headers = self._get_headers()
         payload = self._build_payload(messages, json_schema, creativity, thinking_budget=thinking_budget, force_json=force_json)
 
@@ -1741,6 +1816,8 @@ def _detect_model_source(user_model, model_source):
     still use this name. Will be inlined in a future cleanup."""
     if model_source and model_source.lower() == "claude-code":
         return "claude-code"
+    if model_source and model_source.lower() == "claude-agent":
+        return "claude-agent"
     return detect_provider(user_model, provider=model_source)
 
 

{cat_stack-2.0.1 → cat_stack-2.1.1}/src/catstack/image_functions.py

@@ -1,6 +1,7 @@
 import warnings
 
 from .text_functions import _detect_model_source
+from ._providers import _require_http_provider
 from ._utils import _clean_label, _extract_balanced_json
 from .calls.image_stepback import get_image_stepback_insight
 
@@ -147,6 +148,7 @@ 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")
 
     image_files = _load_image_files(image_input)
 

{cat_stack-2.0.1 → cat_stack-2.1.1}/src/catstack/pdf_functions.py

@@ -1,6 +1,7 @@
 import warnings
 
 from .text_functions import _detect_model_source
+from ._providers import _require_http_provider
 from ._utils import _clean_label, _extract_balanced_json
 from .calls.pdf_stepback import get_pdf_stepback_insight
 
@@ -392,6 +393,7 @@ 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")
 
     # Providers with native PDF support (only used in image/both modes)
     native_pdf_providers = {"anthropic", "google"}

{cat_stack-2.0.1 → cat_stack-2.1.1}/src/catstack/text_functions.py

@@ -410,7 +410,7 @@ def explore_corpus(
     provider = detect_provider(model, provider)
 
     # Validate api_key
-    if provider not in ("ollama", "claude-code") and not api_key:
+    if provider not in ("ollama", "claude-code", "claude-agent") and not api_key:
         raise ValueError(f"api_key is required for provider '{provider}'")
 
     print(f"Exploring categories for question: '{survey_question}'")
@@ -596,7 +596,7 @@ def explore_common_categories(
     provider = detect_provider(model, provider)
 
     # Validate api_key
-    if provider not in ("ollama", "claude-code") and not api_key:
+    if provider not in ("ollama", "claude-code", "claude-agent") and not api_key:
         raise ValueError(f"api_key is required for provider '{provider}'")
 
     # Ollama-specific checks
@@ -1062,7 +1062,7 @@ def multi_class(
     provider = detect_provider(model, provider)
 
     # Validate api_key requirement
-    if provider not in ("ollama", "claude-code") and not api_key:
+    if provider not in ("ollama", "claude-code", "claude-agent") and not api_key:
         raise ValueError(f"api_key is required for provider '{provider}'")
 
     # Handle categories="auto" - auto-detect categories from the data

{cat_stack-2.0.1 → cat_stack-2.1.1}/src/catstack/text_functions_ensemble.py

@@ -660,6 +660,18 @@ def prepare_model_configs(
                     "Install: https://docs.anthropic.com/en/docs/claude-code\n"
                     + "="*60
                 )
+        elif detected_provider == "claude-agent":
+            try:
+                import catclaws  # noqa: F401
+            except ImportError:
+                raise ConnectionError(
+                    "\n" + "="*60 + "\n"
+                    "  CAT-AGENT NOT INSTALLED\n"
+                    "="*60 + "\n\n"
+                    "The cat-claws package is required to use claude-agent as a provider.\n"
+                    "Install: pip install cat-stack[agent]\n"
+                    + "="*60
+                )
         else:
             # Validate API key exists for cloud providers
             if not api_key:
@@ -670,7 +682,7 @@ def prepare_model_configs(
         # Preflight probe: test the model with a minimal JSON call to catch
         # issues (model not found, structured output not supported) before
         # processing thousands of rows.
-        if detected_provider not in ("ollama", "claude-code"):
+        if detected_provider not in ("ollama", "claude-code", "claude-agent"):
             try:
                 probe_client = UnifiedLLMClient(
                     provider=detected_provider,