You are helping me build the following AI agent workflow. ## Goal Multi-Modal Discord Bot With Llama And Gemini: Monitors specific users in Discord to analyze images via Groq, transcribe audio, generate text responses with Llama. ## Specification - What it does: You get real-time analysis of your Discord messages as the bot detects and processes text, images, and audio from specific users. It transcribes your voice notes, analyzes shared photos, and generates thoughtful text responses or new images based on your prompts using Llama and Gemini. - Trigger: Triggered by an event (Event ยท on new Discord message) - Autonomy: Fully hands-off - Expected setup effort: a few focused days - Tools/services involved: - Discord: Message monitoring and response posting - Ollama Chat Model: Llama AI text processing and conversation context - SerpApi (Google Search): Knowledge retrieval for AI responses - Wikipedia: Knowledge base integration ## Known pitfalls, handle each one explicitly in your implementation 1. Large audio files may exceed Groq's transcription size limits so split long recordings into smaller chunks before sending. 2. Ollama models can consume significant RAM on your server so monitor memory usage and set appropriate concurrency limits. 3. Gemini image generation requests can fail silently if the prompt contains restricted content so implement a pre-filtering step to sanitize inputs. ## Reference implementation https://n8n.io/workflows/3222 (workflow template) Fetch and inspect this before building. If it matches my stack, adapt it; if not, rebuild the pattern with my tools. ## Process requirements 1. Before building: ask me which of the listed tools I actually use and what my platform is (n8n / Make / code / other). Do not assume. 2. Adapt the pattern to my answers; do not force the reference stack. 3. Address every pitfall above; tell me how you handled each. 4. Provide a test plan I can run before letting this touch real data. 5. Ask before any step that sends messages, modifies data, or spends money. Source: https://usecasesforagents.com/use-case/multi-modal-discord-bot-with-llama-and-gemini/ via usecasesforagents.com