fastworkflow 2.15.10__tar.gz → 2.17.11__tar.gz

{fastworkflow-2.15.10 → fastworkflow-2.17.11}/PKG-INFO

@@ -1,19 +1,22 @@
 Metadata-Version: 2.1
 Name: fastworkflow
-Version: 2.15.10
+Version: 2.17.11
 Summary: A framework for rapidly building large-scale, deterministic, interactive workflows with a fault-tolerant, conversational UX
 License: Apache-2.0
 Keywords: fastworkflow,ai,workflow,llm,openai
 Author: Dhar Rawal
 Author-email: drawal@radiantlogic.com
-Requires-Python: >=3.11
+Requires-Python: >=3.11,<3.14
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
+Provides-Extra: fastapi
 Requires-Dist: colorama (>=0.4.6,<0.5.0)
 Requires-Dist: datasets (>=4.0.0,<5.0.0)
 Requires-Dist: dspy (>=3.0.1,<4.0.0)
+Requires-Dist: fastapi (>=0.115.5,<0.116.0) ; extra == "fastapi"
+Requires-Dist: fastapi-mcp (>=0.4.0,<0.5.0) ; extra == "fastapi"
 Requires-Dist: libcst (>=1.8.2,<2.0.0)
 Requires-Dist: litellm[proxy] (>=1.75.8,<2.0.0)
 Requires-Dist: mmh3 (>=5.1.0,<6.0.0)
@@ -21,12 +24,14 @@ Requires-Dist: openai (>=1.99.5,<1.100.0)
 Requires-Dist: prompt_toolkit (>=3.0.43,<4.0.0)
 Requires-Dist: pydantic (>=2.9.2,<3.0.0)
 Requires-Dist: python-dotenv (>=1.0.1,<2.0.0)
+Requires-Dist: python-jose[cryptography] (>=3.3.0,<4.0.0) ; extra == "fastapi"
 Requires-Dist: python-levenshtein (>=0.27.1,<0.28.0)
 Requires-Dist: scikit-learn (>=1.6.1,<2.0.0)
 Requires-Dist: sentence-transformers (>=3.4.1,<4.0.0)
 Requires-Dist: speedict (>=0.3.12,<0.4.0)
 Requires-Dist: torch (>=2.7.1,<3.0.0)
 Requires-Dist: transformers (>=4.48.2,<5.0.0)
+Requires-Dist: uvicorn (>=0.29.0,<0.30.0) ; extra == "fastapi"
 Project-URL: homepage, https://github.com/radiantlogicinc/fastworkflow
 Project-URL: repository, https://github.com/radiantlogicinc/fastworkflow
 Description-Content-Type: text/markdown
@@ -581,6 +586,7 @@ This single command will generate the `greet.py` command, `get_properties` and `
 | `LLM_RESPONSE_GEN` | LiteLLM model string for response generation | `run` | `mistral/mistral-small-latest` |
 | `LLM_PLANNER` | LiteLLM model string for the agent's task planner | `run` (agent mode) | `mistral/mistral-small-latest` |
 | `LLM_AGENT` | LiteLLM model string for the DSPy agent | `run` (agent mode) | `mistral/mistral-small-latest` |
+| `LLM_CONVERSATION_STORE` | LiteLLM model string for conversation topic/summary generation | FastAPI service | `mistral/mistral-small-latest` |
 | `NOT_FOUND` | Placeholder value for missing parameters during extraction | Always | `"NOT_FOUND"` |
 | `MISSING_INFORMATION_ERRMSG` | Error message prefix for missing parameters | Always | `"Missing required..."` |
 | `INVALID_INFORMATION_ERRMSG` | Error message prefix for invalid parameters | Always | `"Invalid information..."` |
@@ -594,6 +600,7 @@ This single command will generate the `greet.py` command, `get_properties` and `
 | `LITELLM_API_KEY_RESPONSE_GEN`| API key for the `LLM_RESPONSE_GEN` model | `run` | *required* |
 | `LITELLM_API_KEY_PLANNER`| API key for the `LLM_PLANNER` model | `run` (agent mode) | *required* |
 | `LITELLM_API_KEY_AGENT`| API key for the `LLM_AGENT` model | `run` (agent mode) | *required* |
+| `LITELLM_API_KEY_CONVERSATION_STORE`| API key for the `LLM_CONVERSATION_STORE` model | FastAPI service | *required* |
 
 > [!tip]
 > The example workflows are configured to use Mistral's models by default. You can get a free API key from [Mistral AI](https://mistral.ai) that works with the `mistral-small-latest` model.
@@ -629,7 +636,7 @@ Interested in contributing to `fastWorkflow` itself? Great!
 1.  **Clone the repository:** `git clone https://github.com/your-repo/fastworkflow.git`
 2.  **Set up the environment:** Create a virtual environment using your preferred tool (venv, uv, conda, poetry, etc.) with Python 3.11+
 3.  **Install in editable mode with dev dependencies:** `pip install -e .` or `uv pip install -e ".[dev]"`
-4.  **[Join our Discord](https://discord.gg/CCpNujh2):** Ask questions, discuss functionality, showcase your fastWorkflows
+4.  **[Join our Discord](https://discord.gg/k2g58dDjYR):** Ask questions, discuss functionality, showcase your fastWorkflows
 
 ---
 

{fastworkflow-2.15.10 → fastworkflow-2.17.11}/README.md

@@ -548,6 +548,7 @@ This single command will generate the `greet.py` command, `get_properties` and `
 | `LLM_RESPONSE_GEN` | LiteLLM model string for response generation | `run` | `mistral/mistral-small-latest` |
 | `LLM_PLANNER` | LiteLLM model string for the agent's task planner | `run` (agent mode) | `mistral/mistral-small-latest` |
 | `LLM_AGENT` | LiteLLM model string for the DSPy agent | `run` (agent mode) | `mistral/mistral-small-latest` |
+| `LLM_CONVERSATION_STORE` | LiteLLM model string for conversation topic/summary generation | FastAPI service | `mistral/mistral-small-latest` |
 | `NOT_FOUND` | Placeholder value for missing parameters during extraction | Always | `"NOT_FOUND"` |
 | `MISSING_INFORMATION_ERRMSG` | Error message prefix for missing parameters | Always | `"Missing required..."` |
 | `INVALID_INFORMATION_ERRMSG` | Error message prefix for invalid parameters | Always | `"Invalid information..."` |
@@ -561,6 +562,7 @@ This single command will generate the `greet.py` command, `get_properties` and `
 | `LITELLM_API_KEY_RESPONSE_GEN`| API key for the `LLM_RESPONSE_GEN` model | `run` | *required* |
 | `LITELLM_API_KEY_PLANNER`| API key for the `LLM_PLANNER` model | `run` (agent mode) | *required* |
 | `LITELLM_API_KEY_AGENT`| API key for the `LLM_AGENT` model | `run` (agent mode) | *required* |
+| `LITELLM_API_KEY_CONVERSATION_STORE`| API key for the `LLM_CONVERSATION_STORE` model | FastAPI service | *required* |
 
 > [!tip]
 > The example workflows are configured to use Mistral's models by default. You can get a free API key from [Mistral AI](https://mistral.ai) that works with the `mistral-small-latest` model.
@@ -596,7 +598,7 @@ Interested in contributing to `fastWorkflow` itself? Great!
 1.  **Clone the repository:** `git clone https://github.com/your-repo/fastworkflow.git`
 2.  **Set up the environment:** Create a virtual environment using your preferred tool (venv, uv, conda, poetry, etc.) with Python 3.11+
 3.  **Install in editable mode with dev dependencies:** `pip install -e .` or `uv pip install -e ".[dev]"`
-4.  **[Join our Discord](https://discord.gg/CCpNujh2):** Ask questions, discuss functionality, showcase your fastWorkflows
+4.  **[Join our Discord](https://discord.gg/k2g58dDjYR):** Ask questions, discuss functionality, showcase your fastWorkflows
 
 ---
 

{fastworkflow-2.15.10 → fastworkflow-2.17.11}/fastworkflow/_workflows/command_metadata_extraction/_commands/IntentDetection/what_can_i_do.py

@@ -8,6 +8,7 @@ import json
 
 import fastworkflow
 from fastworkflow.train.generate_synthetic import generate_diverse_utterances
+from fastworkflow.command_context_model import get_workflow_info
 from fastworkflow.command_metadata_api import CommandMetadataAPI
 
 class Signature:
@@ -139,20 +140,33 @@ class ResponseGenerator:
         with contextlib.suppress(Exception):
             if fastworkflow.chat_session:
                 is_agent_mode = fastworkflow.chat_session.run_as_agent
-        
+
         app_workflow = workflow.context["app_workflow"]
-        response = CommandMetadataAPI.get_command_display_text(
+        
+        # Get workflow definition
+        workflow_info = get_workflow_info(app_workflow.folderpath)
+        workflow_def_text = CommandMetadataAPI.get_workflow_definition_display_text(workflow_info)
+        
+        # Get available commands in current context
+        commands_text = CommandMetadataAPI.get_command_display_text(
             subject_workflow_path=app_workflow.folderpath,
             cme_workflow_path=workflow.folderpath,
             active_context_name=app_workflow.current_command_context_name,
             for_agents=is_agent_mode,
         )
+        
+        response = f"{workflow_def_text}\n\n{commands_text}"
 
+        nlu_pipeline_stage = workflow.context.get(
+            "NLU_Pipeline_Stage", 
+            fastworkflow.NLUPipelineStage.INTENT_DETECTION)
+        success = nlu_pipeline_stage == fastworkflow.NLUPipelineStage.INTENT_DETECTION
         return fastworkflow.CommandOutput(
             workflow_id=workflow.id,
             command_responses=[
                 fastworkflow.CommandResponse(
                     response=response,
+                    success=success
                 )
             ]
         ) 

{fastworkflow-2.15.10 → fastworkflow-2.17.11}/fastworkflow/_workflows/command_metadata_extraction/_commands/wildcard.py

@@ -67,7 +67,10 @@ class ResponseGenerator:
             if cnp_output.command_name == 'ErrorCorrection/you_misunderstood':
                 workflow_context["NLU_Pipeline_Stage"] = NLUPipelineStage.INTENT_MISUNDERSTANDING_CLARIFICATION
                 workflow_context["command"] = command
-            else:
+            elif (
+                nlu_pipeline_stage == fastworkflow.NLUPipelineStage.INTENT_DETECTION or
+                cnp_output.command_name == 'ErrorCorrection/abort'
+            ):
                 workflow.end_command_processing()
             workflow.context = workflow_context
 
@@ -76,9 +79,13 @@ class ResponseGenerator:
                 command=command,
             )
             command_output = CommandExecutor.perform_action(workflow, startup_action)
-            command_output.command_responses[0].artifacts["command_handled"] = True     
-            # Set the additional attributes
-            command_output.command_name = cnp_output.command_name
+            if (
+                nlu_pipeline_stage == fastworkflow.NLUPipelineStage.INTENT_DETECTION or
+                cnp_output.command_name == 'ErrorCorrection/abort'
+            ):
+                command_output.command_responses[0].artifacts["command_handled"] = True     
+                # Set the additional attributes
+                command_output.command_name = cnp_output.command_name
             return command_output
         
         if nlu_pipeline_stage in {
@@ -138,7 +145,9 @@ class ResponseGenerator:
             workflow.context = workflow_context
 
         command_name = cnp_output.command_name
-        extractor = ParameterExtraction(workflow, app_workflow, command_name, command)
+        # Use the preserved original command (with parameters) if available
+        preserved_command = f'{command_name}: {workflow.context.get("command", command)}'
+        extractor = ParameterExtraction(workflow, app_workflow, command_name, preserved_command)
         pe_output = extractor.extract()
         if not pe_output.parameters_are_valid:
             return CommandOutput(
@@ -161,7 +170,7 @@ class ResponseGenerator:
                 CommandResponse(
                     response="",
                     artifacts={
-                        "command": command,
+                        "command": preserved_command,
                         "command_name": command_name,
                         "cmd_parameters": pe_output.cmd_parameters,
                     },

{fastworkflow-2.15.10 → fastworkflow-2.17.11}/fastworkflow/_workflows/command_metadata_extraction/intent_detection.py

@@ -96,8 +96,8 @@ class CommandNamePrediction:
                 ].plain_utterances
             }
 
-        # See if the command starts with a command name followed by a space
-        tentative_command_name = command.split(" ", 1)[0]
+        # See if the command starts with a command name followed by a space or a '('
+        tentative_command_name = command.split(" ", 1)[0].split("(", 1)[0]
         normalized_command_name = tentative_command_name.lower()
         command_name = None
         if normalized_command_name in command_name_dict:
@@ -136,7 +136,7 @@ class CommandNamePrediction:
             NLUPipelineStage.INTENT_AMBIGUITY_CLARIFICATION,
             NLUPipelineStage.INTENT_MISUNDERSTANDING_CLARIFICATION
         ) and not command_name:
-            command_name = "what_can_i_do"
+            command_name = "what can i do?"
 
         if not command_name or command_name == "wildcard":
             fully_qualified_command_name=None

{fastworkflow-2.15.10 → fastworkflow-2.17.11}/fastworkflow/_workflows/command_metadata_extraction/parameter_extraction.py

@@ -1,8 +1,10 @@
 import contextlib
 import sys
+import re
 from typing import Dict, List, Optional
 
 from pydantic import BaseModel
+from pydantic_core import PydanticUndefined
 
 import fastworkflow
 from fastworkflow.utils.logging import logger
@@ -59,11 +61,29 @@ class ParameterExtraction:
         if stored_params:
             new_params = self._extract_and_merge_missing_parameters(stored_params, self.command)
         else:
-            # Otherwise use the LLM-based extraction
-            new_params = input_for_param_extraction.extract_parameters(
-                command_parameters_class, 
-                self.command_name, 
-                app_workflow_folderpath)
+            # Check if we're in agentic mode (not assistant mode command)
+            is_agentic_mode = (
+                "is_assistant_mode_command" not in self.cme_workflow.context
+                and "run_as_agent" in self.app_workflow.context
+                and self.app_workflow.context["run_as_agent"]
+            )
+
+            if is_agentic_mode:
+                # Try regex-based extraction first in agentic mode
+                new_params = self._extract_parameters_from_xml(self.command, command_parameters_class)
+
+                # If regex extraction fails, fall back to LLM-based extraction
+                if new_params is None:
+                    new_params = input_for_param_extraction.extract_parameters(
+                        command_parameters_class,
+                        self.command_name,
+                        app_workflow_folderpath)
+            else:
+                # Use LLM-based extraction for assistant mode
+                new_params = input_for_param_extraction.extract_parameters(
+                    command_parameters_class,
+                    self.command_name,
+                    app_workflow_folderpath)
 
         is_valid, error_msg, suggestions, missing_invalid_fields = \
             input_for_param_extraction.validate_parameters(
@@ -272,6 +292,70 @@ class ParameterExtraction:
         # Construct model without validation
         return default_params.__class__.model_construct(**params_data)
 
+    @staticmethod
+    def _extract_parameters_from_xml(command: str, command_parameters_class: type[BaseModel]) -> Optional[BaseModel]:
+        """
+        Extract parameters from XML-formatted command using regex.
+
+        Returns:
+            BaseModel instance with extracted parameters, or None if parsing fails
+        """
+        field_names = list(command_parameters_class.model_fields.keys())
+
+        # If no parameters are defined, return empty model immediately
+        if not field_names:
+            return command_parameters_class.model_construct()
+
+        extracted_data = {}
+
+        # Try to extract each parameter using XML tags
+        if len(field_names) == 1:
+            # If there's only one field, extract content from first XML tag
+            pattern = r'<[^>]+>(.+?)</[^>]+>'
+            if match := re.search(pattern, command, re.DOTALL):
+                parameter_value = match[1].strip()
+                extracted_data[field_names[0]] = parameter_value
+        else:
+            # Try to extract each parameter using XML tags
+            for field_name in field_names:
+                # Look for <field_name>value</field_name> pattern
+                pattern = rf'<{re.escape(field_name)}>(.+?)</{re.escape(field_name)}>'
+                if match := re.search(pattern, command, re.DOTALL):
+                    parameter_value = match[1].strip()
+                    extracted_data[field_name] = parameter_value
+
+        # Check if we extracted values for ALL fields (safest criteria for LLM fallback)
+        all_fields_extracted = len(extracted_data) == len(field_names)
+
+        # Check if agent used example values
+        if all_fields_extracted:
+            for field_name, extracted_value in extracted_data.items():
+                field_info = command_parameters_class.model_fields[field_name]
+                examples = getattr(field_info, "examples", None)
+                if examples and extracted_value in examples:
+                    all_fields_extracted = False
+                    break
+
+        if all_fields_extracted:
+            # Initialize all fields with their default values (if they exist) or None
+            params_data = {}
+            for field_name in field_names:
+                field_info = command_parameters_class.model_fields[field_name]
+                if field_info.default is not PydanticUndefined:
+                    params_data[field_name] = field_info.default
+                elif field_info.default_factory is not None:
+                    params_data[field_name] = field_info.default_factory()
+                else:
+                    params_data[field_name] = None
+
+            # Update with extracted values
+            params_data |= extracted_data
+
+            # Construct model without validation
+            return command_parameters_class.model_construct(**params_data)
+
+        return None
+
     @staticmethod
     def _extract_and_merge_missing_parameters(stored_params: BaseModel, command: str):
         """