improvised langgraph implementaation

README.md

@@ -2,34 +2,35 @@
 
 ## 📋 Overview
 
-FinRyver is an AI-powered financial statement generation platform that automatically converts trial balance data into comprehensive financial reports including balance sheets, cash flow statements, and profit & loss statements. Built with FastAPI and leveraging Large Language Models (LLMs), it streamlines the financial reporting process for accountants, auditors, and financial professionals by automating the generation of detailed financial notes and statements from structured trial balance inputs.
+FinRyver is an AI-powered financial statement generation platform that automatically converts trial balance data into comprehensive financial reports including balance sheets, cash flow statements, and profit & loss statements. Built with FastAPI and leveraging Large Language Models (LLMs) through LangGraph workflows, it streamlines the financial reporting process for accountants, auditors, and financial professionals.
 
-**New in 2025**: FinRyver now features an intelligent **agentic system** powered by LangChain that provides AI-driven automation, natural language processing, and intelligent decision-making for financial statement generation.
+**LangGraph Architecture**: FinRyver now features an intelligent **agentic system** powered by LangGraph that provides AI-driven workflow orchestration, state management, and intelligent task coordination for financial statement generation.
 
 ## 🎯 Key Features
 
 - **Automated Trial Balance Processing**: Upload Excel files containing trial balance data and automatically extract structured financial information
 - **AI-Powered Financial Notes Generation**: Utilize LLMs to generate comprehensive financial notes with detailed explanations and context
-- **Intelligent Agentic System**: LangChain-powered agents that understand natural language instructions and automate complex financial workflows
+- **LangGraph Workflow Orchestration**: State-driven workflows that manage complex financial processing tasks with proper error handling and monitoring
 - **Multi-Statement Support**: Generate Balance Sheets, Cash Flow Statements, and Profit & Loss statements from the same data source
 - **Excel Output Generation**: Export all generated reports and notes to professional Excel formats
 - **RESTful API Architecture**: Easy integration with existing financial systems through well-documented REST endpoints
-- **Flexible Note Selection**: Generate specific financial notes by number or create comprehensive reports covering all relevant sections
-- **Unified Agent Interface**: Single `/agent/generate` endpoint for intelligent financial statement generation
+- **Specialized Endpoints**: Dedicated routes for each financial statement type (`/notes`, `/pnl`, `/bs`, `/cf`)
+- **Performance Monitoring**: Built-in timing and status tracking for all agentic workflows
 
 ## 🏗️ Project Architecture
 
 ```
 FinRyver/
-├── app.py                 # Main FastAPI application with API endpoints
-├── requirements.txt       # Python dependencies
+├── app.py                 # Main FastAPI application with 4 specialized routes
+├── requirements.txt       # Python dependencies 
 ├── Dockerfile            # Container configuration
 ├── docker-compose.yml    # Multi-container orchestration
 ├── 
-├── agents/               # Agentic system (LangChain)
-│   ├── base_config.py    # Agent configuration and utilities
-│   ├── simple_agent.py   # Financial statement agent
-│   └── simple_tools.py   # LangChain tools for financial processing
+├── agents/               # LangGraph-based agentic system
+│   ├── langgraph.py   # LangGraph workflow definitions
+│   ├── simple_tools.py       # LangChain tools for financial processing
+│   ├── base_config.py        # Agent configuration and utilities
+│   └── simple_agent.py       # Financial statement agent (legacy)
 ├── 
 ├── bs/                   # Balance Sheet processing modules
 │   ├── bl_llm.py         # Balance sheet LLM integration
@@ -89,7 +90,8 @@ Trial Balance Upload → Data Extraction → AI Processing → Financial Stateme
 
 ### AI/ML Integration
 - **Large Language Models (LLMs)**: For intelligent financial note generation and analysis
-- **LangChain Framework**: Agentic system for intelligent financial statement processing
+- **LangChain Framework**: Tool integration and AI agent development
+- **LangGraph**: Workflow orchestration and state management for complex financial processing
 - **OpenRouter API**: Flexible LLM provider integration for AI-powered analysis
 - **Custom AI Pipelines**: Specialized processing for financial data interpretation
 
@@ -134,31 +136,51 @@ Trial Balance Upload → Data Extraction → AI Processing → Financial Stateme
 
 | Endpoint | Method | Description |
 |----------|--------|-------------|
-| `/new` | POST | Generate financial notes and Excel output from trial balance |
-| `/hardcoded` | POST | Process predefined trial balance templates |
-| `/bs_from_notes` | POST | Generate balance sheet from existing notes |
-| `/pnl_from_notes` | POST | Generate P&L statement from existing notes |
-| `/cf_from_notes` | POST | Generate cash flow statement from existing notes |
-| `/agent/generate` | POST | **NEW**: Intelligent agent-based financial statement generation |
+| `/notes` | POST | Generate financial notes from trial balance using LangGraph workflow |
+| `/pnl` | POST | Generate P&L statement from trial balance using LangGraph workflow |
+| `/bs` | POST | Generate balance sheet from trial balance using LangGraph workflow |
+| `/cf` | POST | Generate cash flow statement from trial balance using LangGraph workflow |
 
-#### Agentic System Endpoint
+#### LangGraph-Powered Endpoints
 
-**`POST /agent/generate`** - Unified intelligent financial statement generation
+All endpoints follow the same pattern and use LangGraph workflows for intelligent task orchestration:
 
 **Parameters:**
-- `file`: Trial balance Excel file
-- `note_numbers`: Optional comma-separated note numbers (empty = all notes)
-- `statement_type`: "all", "notes", "balance_sheet", "pnl", or "cash_flow"
+- `file`: Trial balance Excel file (multipart/form-data)
+
+**Response:**
+- Excel file download with the generated financial statement
 
 **Example Usage:**
 ```bash
-curl -X POST "http://localhost:8000/agent/generate" \
+# Generate financial notes
+curl -X POST "http://localhost:8000/notes" \
+  -H "Content-Type: multipart/form-data" \
+  --output notes.xlsx
+
+# Generate P&L statement
+curl -X POST "http://localhost:8000/pnl" \
+  -H "Content-Type: multipart/form-data" \
+  --output pnl_statement.xlsx
+
+# Generate balance sheet
+curl -X POST "http://localhost:8000/bs" \
+  -H "Content-Type: multipart/form-data" \
+  --output balance_sheet.xlsx
+
+# Generate cash flow statement
+curl -X POST "http://localhost:8000/cf" \
   -H "Content-Type: multipart/form-data" \
   -F "file=@trial_balance.xlsx" \
-  -F "note_numbers=2,3,4,5" \
-  -F "statement_type=all"
+  --output cash_flow.xlsx
 ```
 
+**LangGraph Workflow Features:**
+- **State Management**: Each workflow tracks execution state, timing, and errors
+- **Error Handling**: Comprehensive error capture and reporting
+- **Performance Monitoring**: Built-in timing for workflow execution
+- **Tool Integration**: Seamless integration with LangChain tools
+
 ## 📊 Results & Examples
 
 ### Input Format

app.py

@@ -1,21 +1,9 @@
-from fastapi import FastAPI, APIRouter, UploadFile, File, Form, HTTPException
-from fastapi.responses import JSONResponse, PlainTextResponse, FileResponse
-from typing import Optional, Dict, Any
-import pandas as pd
+from fastapi import FastAPI, APIRouter, UploadFile, File, HTTPException
+from fastapi.responses import FileResponse
 import os
 import shutil
-import json
-import subprocess
 import logging
-
-# Import utilities and logic from modular files
-from utils.utils import clean_value
-from notes.data_extraction import extract_trial_balance_data, analyze_and_save_results
-from notes.llm_notes_generator import FlexibleFinancialNoteGenerator
-from notes.notes_generator import process_json
-from notes.json_to_excel import json_to_xlsx
-from utils.utils_normalize import normalize_llm_note_json, normalize_llm_notes_json
-from agents.langgraph_routes import run_workflow
+from agents.langgraph import run_workflow
 
 # Configure logging for the application
 logging.basicConfig(level=logging.INFO)
@@ -35,23 +23,10 @@ async def startup_event():
 @app.on_event("shutdown")
 async def shutdown_event():
     logger.info("Financial Notes Generator API is shutting down.")
-router = APIRouter()
 
-def process_uploaded_file(file: UploadFile) -> pd.DataFrame:
-    os.makedirs("data/input", exist_ok=True)
-    file_location = f"data/input/{file.filename}"
-    with open(file_location, "wb") as buffer:
-        shutil.copyfileobj(file.file, buffer)
-    structured_data = extract_trial_balance_data(file_location)
-    output_file = "data/output1/parsed_trial_balance.json"
-    analyze_and_save_results(structured_data, output_file)
-    with open(output_file, "r", encoding="utf-8") as f:
-        parsed_data = json.load(f)
-    tb_df = pd.DataFrame(parsed_data if isinstance(parsed_data, list) else parsed_data.get("trial_balance", parsed_data))
-    tb_df['amount'] = tb_df['amount'].apply(clean_value)
-    return tb_df
+router = APIRouter()
 
-@router.post("/new")
+"""@router.post("/new")
 async def llm_generate_and_excel(
     file: UploadFile = File(...),
     note_number: Optional[str] = Form(None)
@@ -102,302 +77,8 @@ async def llm_generate_and_excel(
         excel_path,
         filename=os.path.basename(excel_path),
         media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
-    )
-
-@router.post("/hardcoded")
-async def run_full_pipeline(
-    file: UploadFile = File(...),
-    note_number: Optional[str] = Form(None)
-):
-    os.makedirs("data/input", exist_ok=True)
-    file_location = f"data/input/{file.filename}"
-    with open(file_location, "wb") as buffer:
-        shutil.copyfileobj(file.file, buffer)
-    os.makedirs("data/output1", exist_ok=True)
-    structured_data = extract_trial_balance_data(file_location)
-    output1_json = "data/output1/parsed_trial_balance.json"
-    analyze_and_save_results(structured_data, output1_json)
-    os.makedirs("data/output2", exist_ok=True)
-    try:
-        process_json(output1_json)
-    except ImportError:
-        logger.error("main16_23.process_json not found. Please ensure 'app/main16_23.py' exists and is named correctly.")
-        raise HTTPException(status_code=500, detail="main16_23.process_json not found. Please ensure 'app/main16_23.py' exists and is named correctly.")
-    except Exception as e:
-        logger.error(f"main16_23.process_json failed: {e}")
-        raise HTTPException(status_code=500, detail=f"main16_23.process_json failed: {e}")
-    notes_json = "data/output2/notes_output.json"
-    with open(notes_json, "r", encoding="utf-8") as f:
-        notes_data = json.load(f)
-    if isinstance(notes_data, dict):
-        for key in ["notes", "trial_balance"]:
-            if key in notes_data:
-                notes_data = notes_data[key]
-                break
-    def wrap_notes(notes):
-        return {"notes": notes}
-    if note_number:
-        numbers = [n.strip() for n in note_number.split(",")]
-        notes_data = [
-            note for note in notes_data
-            if str(note.get('note_number', '')).strip() in numbers
-        ]
-        filtered_json = "data/output2/notes_output_filtered.json"
-        with open(filtered_json, "w", encoding="utf-8") as f2:
-            json.dump(wrap_notes(notes_data), f2, ensure_ascii=False, indent=2)
-        json_input_for_excel = filtered_json
-    else:
-        temp_json = "data/output2/notes_output_wrapped.json"
-        with open(temp_json, "w", encoding="utf-8") as f2:
-            json.dump(wrap_notes(notes_data), f2, ensure_ascii=False, indent=2)
-        json_input_for_excel = temp_json
-    os.makedirs("data/output3", exist_ok=True)
-    try:
-        output3_xlsx = "data/output3/final_output.xlsx"
-        json_to_xlsx(json_input_for_excel, output3_xlsx)
-    except ImportError:
-        logger.error("json_xlsx.json_to_xlsx not found")
-        raise HTTPException(status_code=500, detail="json_xlsx.json_to_xlsx not found")
-    except Exception as e:
-        logger.error(f"json_xlsx.json_to_xlsx failed: {e}")
-        raise HTTPException(status_code=500, detail=f"json_xlsx.json_to_xlsx failed: {e}")
-    return FileResponse(
-        output3_xlsx,
-        filename=os.path.basename(output3_xlsx),
-        media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
-    )
-
-def run_subprocess(
-    script_path: str,
-    args: list,
-    env: Dict[str, str],
-    cwd: str
-) -> subprocess.CompletedProcess:
-    try:
-        logger.info(f"Running {script_path} with args {args} in {cwd}")
-        result = subprocess.run(
-            ["python", script_path] + args,
-            capture_output=True,
-            text=True,
-            check=True,
-            env=env,
-            cwd=cwd
-        )
-        logger.debug(f"{script_path} STDOUT:\n{result.stdout}")
-        logger.debug(f"{script_path} STDERR:\n{result.stderr}")
-        return result
-    except subprocess.CalledProcessError as e:
-        logger.error(f"{script_path} failed: {e}")
-        logger.error(f"STDOUT: {e.stdout}")
-        logger.error(f"STDERR: {e.stderr}")
-        raise HTTPException(
-            status_code=500,
-            detail=f"{script_path} failed: {e}\nSTDOUT:\n{e.stdout}\nSTDERR:\n{e.stderr}"
-        )
-
-def extract_output_file(stdout: str, keyword: str = "Output file:") -> Optional[str]:
-    for line in stdout.splitlines():
-        if keyword in line:
-            return line.split(keyword)[-1].strip()
-    return None
-
-@router.post("/bs_from_notes")
-async def bs_from_notes(file: UploadFile = File(...)):
-    os.makedirs("data/input", exist_ok=True)
-    input_excel_path = os.path.join("data/input", file.filename)
-    with open(input_excel_path, "wb") as buffer:
-        shutil.copyfileobj(file.file, buffer)
-    logger.info(f"Uploaded Excel saved to: {input_excel_path}")
-    logger.info(f"Files in data/input/: {os.listdir('data/input')}")
-    env = os.environ.copy()
-    if os.getenv("OPENROUTER_API_KEY"):
-        env["OPENROUTER_API_KEY"] = os.getenv("OPENROUTER_API_KEY")
-    env["INPUT_FILE"] = "data/clean_financial_data_bs.json"
-    cwd = os.getenv("PROJECT_ROOT", os.getcwd())
-    # Run Balance Sheet Data Extractor
-    run_subprocess("bs/balance_sheet_data_extractor.py", [input_excel_path], env, cwd)
-    logger.info(f"Files in data/csv_notes_bs/: {os.listdir('data/csv_notes_bs') if os.path.exists('data/csv_notes_bs') else 'data/csv_notes_bs does not exist'}")
-    # Run Balance Sheet CSV to JSON Converter
-    run_subprocess("bs/balance_sheet_csv_to_json_converter.py", [], env, cwd)
-    logger.info(f"data/clean_financial_data_bs.json exists: {os.path.exists('data/clean_financial_data_bs.json')}")
-    # Run Balance Sheet Generator
-    result = run_subprocess("bs/balance_sheet_generator.py", [], env, cwd)
-    output_file = extract_output_file(result.stdout)
-    if output_file and not os.path.isabs(output_file):
-        output_file_path = os.path.join(cwd, output_file)
-    else:
-        output_file_path = output_file
-    if not output_file or not os.path.exists(output_file_path):
-        debug_msg = f"\nSTDOUT:\n{result.stdout}\nSTDERR:\n{result.stderr}"
-        logger.error(f"Could not determine output file from balance_sheet_generator.py output.{debug_msg}")
-        raise HTTPException(status_code=500, detail=f"Could not determine output file from balance_sheet_generator.py output.{debug_msg}")
-    logger.info(f"Pipeline completed. Output file: {output_file_path}")
-    return FileResponse(
-        output_file_path,
-        filename=os.path.basename(output_file_path),
-        media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
-    )
-
-@router.post("/pnl_from_notes")
-async def pnl_from_notes(file: UploadFile = File(...)):
-    os.makedirs("data/input", exist_ok=True)
-    input_excel_path = os.path.join("data/input", file.filename)
-    with open(input_excel_path, "wb") as buffer:
-        shutil.copyfileobj(file.file, buffer)
-    logger.info(f"Uploaded Excel saved to: {input_excel_path}")
-    logger.info(f"Files in data/input/: {os.listdir('data/input')}")
-    env = os.environ.copy()
-    if os.getenv("OPENROUTER_API_KEY"):
-        env["OPENROUTER_API_KEY"] = os.getenv("OPENROUTER_API_KEY")
-    env["INPUT_FILE"] = "data/clean_financial_data_pnl.json"
-    cwd = os.getenv("PROJECT_ROOT", os.getcwd())
-    # Run Profit & Loss Data Extractor
-    run_subprocess("pnl/profit_loss_data_extractor.py", [input_excel_path], env, cwd)
-    csv_notes_pnl_path = os.path.join(cwd, 'data/csv_notes_pnl')
-    logger.info(f"Files in {csv_notes_pnl_path}/: {os.listdir(csv_notes_pnl_path) if os.path.exists(csv_notes_pnl_path) else f'{csv_notes_pnl_path} does not exist'}")
-    # Run Profit & Loss CSV to JSON Converter
-    run_subprocess("pnl/profit_loss_csv_to_json_converter.py", [], env, cwd)
-    json_path = os.path.join(cwd, 'data/clean_financial_data_pnl.json')
-    logger.info(f"data/clean_financial_data_pnl.json exists: {os.path.exists(json_path)}")
-    # Run Profit & Loss Statement Generator
-    run_subprocess("pnl/profit_loss_statement_generator.py", [], env, cwd)
-    # Use fixed output file path
-    output_file_path = os.path.join(cwd, "data/pnl_statement.xlsx")
-    if not os.path.exists(output_file_path):
-        logger.error(f"Could not find expected output file for P&L statement: {output_file_path}")
-        raise HTTPException(status_code=500, detail=f"Could not find expected output file for P&L statement: {output_file_path}")
-    logger.info(f"Pipeline completed. Output file: {output_file_path}")
-    return FileResponse(
-        output_file_path,
-        filename=os.path.basename(output_file_path),
-        media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
-    )
-
-@router.post("/cf_from_notes")
-async def cf_from_notes(file: UploadFile = File(...)):
-    os.makedirs("data/input", exist_ok=True)
-    input_excel_path = os.path.join("data/input", file.filename)
-    with open(input_excel_path, "wb") as buffer:
-        shutil.copyfileobj(file.file, buffer)
-    logger.info(f"Uploaded Excel saved to: {input_excel_path}")
-    logger.info(f"Files in data/input/: {os.listdir('data/input')}")
-    env = os.environ.copy()
-    cwd = os.getenv("PROJECT_ROOT", os.getcwd())
-    # Step 1: Run Cash Flow Data Extractor
-    run_subprocess("cf/cash_flow_data_extractor.py", [input_excel_path], env, cwd)
-    csv_notes_cfs_path = os.path.join(cwd, 'data/csv_notes_cfs')
-    logger.info(f"Files in {csv_notes_cfs_path}/: {os.listdir(csv_notes_cfs_path) if os.path.exists(csv_notes_cfs_path) else f'{csv_notes_cfs_path} does not exist'}")
-    # Step 2: Run Cash Flow CSV to JSON Converter
-    run_subprocess("cf/cash_flow_csv_to_json_converter.py", [], env, cwd)
-    json_path = os.path.join(cwd, 'data/clean_financial_data_cfs.json')
-    logger.info(f"data/clean_financial_data_cfs.json exists: {os.path.exists(json_path)}")
-    # Step 3: Run Cash Flow Data Processor
-    run_subprocess("cf/cash_flow_data_processor.py", [], env, cwd)
-    extracted_json_path = os.path.join(cwd, 'data/extracted_cfs_data.json')
-    logger.info(f"data/extracted_cfs_data.json exists: {os.path.exists(extracted_json_path)}")
-    # Step 4: Run Cash Flow Statement Generator
-    result = run_subprocess("cf/cash_flow_statement_generator.py", [], env, cwd)
-    output_file = "data/cash_flow_statements.xlsx"
-    output_file_path = os.path.join(cwd, output_file)
-    if not os.path.exists(output_file_path):
-        output_file_path = os.path.join(cwd, "data/cash_flow_statements.xlsx")
-    if not os.path.exists(output_file_path):
-        debug_msg = f"\nSTDOUT:\n{result.stdout}\nSTDERR:\n{result.stderr}"
-        logger.error(f"Could not determine output file from cf_generation.py output.{debug_msg}")
-        raise HTTPException(status_code=500, detail=f"Could not determine output file from cf_generation.py output.{debug_msg}")
-    logger.info(f"Pipeline completed. Output file: {output_file_path}")
-    return FileResponse(
-        output_file_path,
-        filename=os.path.basename(output_file_path),
-        media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
-    )
-
-@router.post("/agent/generate")
-async def agent_generate_statements(
-    file: UploadFile = File(...),
-    note_numbers: str = Form(""),
-    statement_type: str = Form("notes")  # notes, balance_sheet, pnl, cash_flow only
-):
-    """Unified intelligent generation endpoint.
-
-    Behaviors:
-    - balance_sheet / pnl / cash_flow: run respective tool directly and return the Excel file
-    """
-    try:
-        # Validate statement_type
-        valid_types = {"balance_sheet", "pnl", "cash_flow", "notes"}
-        if statement_type not in valid_types:
-            raise HTTPException(status_code=400, detail=f"Invalid statement_type: {statement_type}. Allowed: balance_sheet, pnl, cash_flow, notes")
+    )"""
 
-        # Persist uploaded file
-        upload_dir = "data/input"
-        os.makedirs(upload_dir, exist_ok=True)
-        file_path = os.path.join(upload_dir, file.filename)
-        with open(file_path, "wb") as buffer:
-            shutil.copyfileobj(file.file, buffer)
-
-        # Direct tool imports (lazy to avoid import cost if not needed)
-        from agents.simple_tools import (
-            generate_notes_full_pipeline_from_path,
-            generate_balance_sheet,
-            generate_pnl_statement,
-            generate_cash_flow_statement,
-        )
-
-        # Single statement short‑circuit (return actual file)
-        if statement_type in {"balance_sheet", "pnl", "cash_flow", "notes"}:
-            if statement_type == "notes":
-                output = generate_notes_full_pipeline_from_path(file_path, note_numbers)
-                if output["status"] != "success":
-                    logger.error(f"Notes generation pipeline failed: {output.get('error')}")
-                    raise HTTPException(status_code=500, detail=f"Notes generation pipeline failed: {output.get('error')}")
-                output3_xlsx = output["output_xlsx_path"]
-                return FileResponse(
-                    output3_xlsx,
-                    filename=os.path.basename(output3_xlsx),
-                    media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
-                )
-
-            if statement_type == "balance_sheet":
-                bs_result = generate_balance_sheet.run({"file_path": file_path}) if hasattr(generate_balance_sheet, "run") else generate_balance_sheet(file_path)
-                if bs_result.get("status") != "success":
-                    raise HTTPException(status_code=500, detail=f"Balance sheet generation failed: {bs_result.get('error')}")
-                # Expect directory with XLSX files
-                output_dir = bs_result.get("output_path", "data/output/")
-                # Choose first xlsx file
-                if os.path.isdir(output_dir):
-                    xlsx_files = [f for f in os.listdir(output_dir) if f.endswith('.xlsx')]
-                    if not xlsx_files:
-                        raise HTTPException(status_code=500, detail="No balance sheet Excel file produced")
-                    output_file = os.path.join(output_dir, xlsx_files[0])
-                else:
-                    output_file = output_dir
-                return FileResponse(output_file, filename=os.path.basename(output_file), media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
-
-            if statement_type == "pnl":
-                pnl_result = generate_pnl_statement.run({"file_path": file_path}) if hasattr(generate_pnl_statement, "run") else generate_pnl_statement(file_path)
-                if pnl_result.get("status") != "success":
-                    raise HTTPException(status_code=500, detail=f"P&L generation failed: {pnl_result.get('error')}")
-                output_path = pnl_result.get("output_path", "data/pnl_statement.xlsx")
-                if not os.path.exists(output_path):
-                    raise HTTPException(status_code=500, detail="P&L Excel file not found")
-                return FileResponse(output_path, filename=os.path.basename(output_path), media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
-
-            if statement_type == "cash_flow":
-                cf_result = generate_cash_flow_statement.run({"file_path": file_path}) if hasattr(generate_cash_flow_statement, "run") else generate_cash_flow_statement(file_path)
-                if cf_result.get("status") != "success":
-                    raise HTTPException(status_code=500, detail=f"Cash flow generation failed: {cf_result.get('error')}")
-                output_path = cf_result.get("output_path", "data/cash_flow_statements.xlsx")
-                if not os.path.exists(output_path):
-                    raise HTTPException(status_code=500, detail="Cash flow Excel file not found")
-                return FileResponse(output_path, filename=os.path.basename(output_path), media_type="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
-
-        
-    except HTTPException:
-        raise
-    except Exception as e:
-        logger.error(f"Error in agent statement generation: {e}")
-        raise HTTPException(status_code=500, detail=f"Agent generation failed: {str(e)}")
 
 @router.post("/notes")
 async def notes_route(file: UploadFile = File(...)):

requirements.txt

@@ -13,4 +13,4 @@ langchain
 langchain-openai
 langchain-community
 langchain-core
-langsmithz
+langgraph