> ## Documentation Index > Fetch the complete documentation index at: https://docs.praison.ai/llms.txt > Use this file to discover all available pages before exploring further. # Agent Recipes Troubleshooting > Diagnose and fix common Agent Recipes issues # Agent Recipes Troubleshooting Common issues and solutions when working with Agent Recipes. ## Diagnostic Commands ### Check All Dependencies ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Check specific recipe praison recipes doctor ai-subtitle-generator # Example output: # Checking dependencies for: ai-subtitle-generator # -------------------------------------------------- # ✓ env:OPENAI_API_KEY # ✓ package:openai # ✗ external:ffmpeg # ✗ tool:whisper_tool # -------------------------------------------------- # Missing dependencies. Install with: # brew install ffmpeg # macOS # apt install ffmpeg # Ubuntu ``` ### Verify API Keys ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Test LLM connectivity python3 -c " import os os.environ['OPENAI_API_KEY'] = 'your-key' from praisonai_tools.recipe_tools import LLMTool llm = LLMTool(provider='openai') print(llm.check_dependencies()) response = llm.complete('Say hello', max_tokens=10) print(f'Success: {response.content}') " ``` ### List Available Recipes ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} praison recipes list --json | python3 -c " import json, sys recipes = json.load(sys.stdin) print(f'Total recipes: {len(recipes)}') for r in recipes[:5]: print(f\" - {r['name']}: {r['description'][:50]}...\") " ``` ## Common Issues ### 1. Missing API Key **Error:** ``` DependencyError: OPENAI_API_KEY not set ``` **Solution:** ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Set environment variable export OPENAI_API_KEY=sk-proj-... # Or in Python import os os.environ['OPENAI_API_KEY'] = 'sk-proj-...' # Verify echo $OPENAI_API_KEY | tail -c 5 # Shows last 4 chars ``` ### 2. Missing External Tool (ffmpeg) **Error:** ``` ✗ external:ffmpeg FileNotFoundError: ffmpeg not found ``` **Solution:** ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # macOS brew install ffmpeg # Ubuntu/Debian sudo apt update && sudo apt install ffmpeg # Windows (with chocolatey) choco install ffmpeg # Verify ffmpeg -version ``` ### 3. Missing External Tool (tesseract) **Error:** ``` ✗ external:tesseract ``` **Solution:** ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # macOS brew install tesseract # Ubuntu/Debian sudo apt install tesseract-ocr # Verify tesseract --version ``` ### 4. Missing Python Package **Error:** ``` ✗ package:openai ModuleNotFoundError: No module named 'openai' ``` **Solution:** ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Install specific package pip install openai # Install all recipe tools pip install praisonai-tools[all] # Install cluster-specific extras pip install praisonai-tools[media] # Video/audio pip install praisonai-tools[image] # Image processing pip install praisonai-tools[docs] # Document processing ``` ### 5. API Rate Limit **Error:** ``` RateLimitError: Rate limit exceeded ``` **Solution:** ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} import time from praisonai_tools.recipe_tools import LLMTool llm = LLMTool(provider="openai") def complete_with_retry(prompt, max_retries=3): for attempt in range(max_retries): try: return llm.complete(prompt) except Exception as e: if "rate_limit" in str(e).lower(): wait = 2 ** attempt # Exponential backoff print(f"Rate limited, waiting {wait}s...") time.sleep(wait) else: raise raise Exception("Max retries exceeded") ``` ### 6. Invalid Model Name **Error:** ``` InvalidRequestError: The model 'gpt-5' does not exist ``` **Solution:** ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Use valid model names from praisonai_tools.recipe_tools import LLMTool # OpenAI models llm = LLMTool(provider="openai", model="gpt-4o-mini") # ✓ llm = LLMTool(provider="openai", model="gpt-4o") # ✓ llm = LLMTool(provider="openai", model="gpt-4-turbo") # ✓ # Anthropic models llm = LLMTool(provider="anthropic", model="claude-3-haiku-20240307") # ✓ llm = LLMTool(provider="anthropic", model="claude-3-sonnet-20240229") # ✓ # Google models llm = LLMTool(provider="google", model="gemini-1.5-flash") # ✓ llm = LLMTool(provider="google", model="gemini-1.5-pro") # ✓ ``` ### 7. Recipe Not Found **Error:** ``` Recipe not found: ai-custom-recipe ``` **Solution:** ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # List available recipes praison recipes list | grep -i "custom" # Check recipe paths ls ~/.praisonai/templates/ ls ~/.config/praison/templates/ ls /Users/praison/Agent-Recipes/agent_recipes/templates/ # Initialize a recipe locally praison recipes init ai-blog-generator --output ./my-recipe ``` ### 8. Permission Denied **Error:** ``` PermissionError: [Errno 13] Permission denied: '/output/file.txt' ``` **Solution:** ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Check directory permissions ls -la ./output/ # Create output directory with proper permissions mkdir -p ./output chmod 755 ./output # Or specify writable output path praison recipes run ai-blog-generator --output ~/Documents/output/ ``` ### 9. Memory Error with Large Files **Error:** ``` MemoryError: Unable to allocate array ``` **Solution:** ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Process large files in chunks import pandas as pd # Instead of loading entire file # df = pd.read_csv("large_file.csv") # ✗ # Use chunked processing chunks = pd.read_csv("large_file.csv", chunksize=10000) for chunk in chunks: # Process each chunk pass ``` ### 10. Timeout Error **Error:** ``` TimeoutError: Request timed out after 30 seconds ``` **Solution:** ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} from praisonai_tools.recipe_tools import LLMTool # Increase timeout llm = LLMTool(provider="openai", timeout=120) # Or use streaming for long responses for chunk in llm.stream("Write a long essay..."): print(chunk, end="", flush=True) ``` ## Debug Mode ### Enable Verbose Logging ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} import logging # Enable debug logging logging.basicConfig(level=logging.DEBUG) # Or for specific module logging.getLogger("praisonai_tools").setLevel(logging.DEBUG) ``` ### Inspect Recipe Configuration ```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} import yaml def inspect_recipe(name): """Print recipe configuration for debugging.""" from pathlib import Path paths = [ Path.home() / ".praison" / "templates" / name, Path("/Users/praison/Agent-Recipes/agent_recipes/templates") / name, ] for path in paths: template = path / "TEMPLATE.yaml" if template.exists(): with open(template) as f: config = yaml.safe_load(f) print(f"Recipe: {name}") print(f"Path: {path}") print(f"Config: {yaml.dump(config, default_flow_style=False)}") return config print(f"Recipe not found: {name}") return None inspect_recipe("ai-blog-generator") ``` ## Getting Help ### Check Documentation ```bash theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}} # Recipe info praison recipes info ai-blog-generator # Recipe explanation praison recipes explain ai-blog-generator ``` ### Report Issues If you encounter a bug: 1. Check existing issues: [GitHub Issues](https://github.com/MervinPraison/PraisonAI/issues) 2. Include: * Recipe name and version * Full error message * Python version (`python --version`) * OS and version * Steps to reproduce ### Community Support * [GitHub Discussions](https://github.com/MervinPraison/PraisonAI/discussions) * [Documentation](https://docs.praisonai.com) ## Next Steps * [Agent Recipes Overview](/docs/examples/agent-recipes) - Getting started * [CLI Reference](/docs/examples/agent-recipes-cli) - Command reference * [Python Integration](/docs/examples/agent-recipes-integration) - Use in applications