HumationHuman + Automation
Back to Catalog

Automatically optimize AI prompts with OpenAI using OPRO & DSPy methodology

Shun NakayamaShun Nakayama
263 views
2/3/2026
Data ExportShopifyAPICSVInventory
Official Page

This workflow implements cutting-edge concepts from Google DeepMind's OPRO (Optimization by PROmpting) and Stanford's DSPy to automatically refine AI prompts. It iteratively generates, evaluates, and optimizes responses against a ground truth, allowing you to "compile" your prompts for maximum accuracy.

Why this is powerful

Instead of manually tweaking prompts (trial and error), this workflow treats prompt engineering as an optimization problem:

  • OPRO-style Optimization: The "Optimizer" LLM analyzes past performance scores and reasons to mathematically deduce a better prompt.
  • DSPy-style Logic: It separates the "Logic" (Workflow) from the "Parameters" (Prompts), allowing the system to self-correct until it matches the Ground Truth.

How it works

  • Define: Set your initial prompt and a test case with the expected answer (Ground Truth).
  • Generate: The workflow generates a response using the current prompt.
  • Evaluate: An AI Evaluator scores the response (0-100) based on accuracy and format.
  • Optimize: If the score is low, the Optimizer AI analyzes the failure and rewrites the prompt.
  • Loop: The process repeats until the score reaches 95/100 or the loop limit is hit.

Setup steps

  1. Configure OpenAI: Ensure you have an OpenAI credential set up in the OpenAI Chat Model node.
  2. Customize: Open the Define Initial Prompt & Test Data node and set your initial_prompt, test_input, and ground_truth.
  3. Run: Execute the workflow and check the Manage Loop & State node output for the optimized prompt.

Automatically Optimize AI Prompts with OpenAI using OPRO + DSPY Methodology

This n8n workflow demonstrates a foundational structure for optimizing AI prompts using an iterative approach, potentially leveraging OPRO (Optimization by Prompting) and DSPY methodologies. While the provided JSON defines a basic flow, it sets the stage for more advanced prompt engineering techniques.

What it does

This workflow currently provides a basic framework for processing and potentially optimizing AI prompts. It includes:

  1. Manual Trigger: Initiates the workflow upon a manual click, allowing for on-demand execution.
  2. AI Agent: A placeholder for an AI agent, likely intended to interact with a Language Model for prompt generation, evaluation, or optimization tasks.
  3. OpenAI Chat Model: Specifies the use of an OpenAI Chat Model, indicating that the workflow is designed to interact with OpenAI's large language models.
  4. Structured Output Parser: A node to parse structured output, suggesting that the AI agent's responses are expected to conform to a specific format (e.g., JSON).
  5. Edit Fields (Set): A node to manipulate or set data fields, useful for preparing input for the AI agent or processing its output.
  6. If Node: Implements conditional logic, allowing the workflow to branch based on specific criteria (e.g., evaluating prompt effectiveness or response quality).
  7. No Operation, do nothing: A placeholder node, indicating a path where no action is currently defined, often used for debugging or future expansion.
  8. Sticky Note: Provides documentation within the workflow itself, explaining the purpose of different sections.

Prerequisites/Requirements

  • n8n Instance: A running n8n instance (self-hosted or cloud).
  • OpenAI API Key: Required for the "OpenAI Chat Model" node to interact with OpenAI's services. This should be configured as an n8n credential.

Setup/Usage

  1. Import the workflow: Download the provided JSON and import it into your n8n instance.
  2. Configure Credentials:
    • Locate the "OpenAI Chat Model" node.
    • Configure your OpenAI API Key as a credential within n8n and select it for this node.
  3. Customize the AI Agent:
    • The "AI Agent" node is currently a placeholder. You will need to configure its specific capabilities, tools, and instructions to implement your desired prompt optimization logic (e.g., OPRO, DSPY).
  4. Define Output Parsing:
    • Configure the "Structured Output Parser" node with the expected schema or format of the output from your AI agent.
  5. Implement Conditional Logic:
    • Modify the "If" node with the conditions necessary to evaluate the AI's responses or prompt effectiveness. This might involve checking for keywords, sentiment, or adherence to specific criteria.
  6. Develop Actionable Paths:
    • Connect the "True" and "False" outputs of the "If" node to further actions. For example, the "True" path might lead to saving an optimized prompt, while the "False" path might trigger another iteration of the AI agent with refined instructions.
  7. Execute the workflow: Click the "Execute workflow" button on the "Manual Trigger" node to run the workflow. You can also activate it to run automatically based on other triggers (e.g., a webhook or schedule) once fully configured.

Related Templates

Auto-create TikTok videos with VEED.io AI avatars, ElevenLabs & GPT-4

💥 Viral TikTok Video Machine: Auto-Create Videos with Your AI Avatar --- 🎯 Who is this for? This workflow is for content creators, marketers, and agencies who want to use Veed.io’s AI avatar technology to produce short, engaging TikTok videos automatically. It’s ideal for creators who want to appear on camera without recording themselves, and for teams managing multiple brands who need to generate videos at scale. --- ⚙️ What problem this workflow solves Manually creating videos for TikTok can take hours — finding trends, writing scripts, recording, and editing. By combining Veed.io, ElevenLabs, and GPT-4, this workflow transforms a simple Telegram input into a ready-to-post TikTok video featuring your AI avatar powered by Veed.io — speaking naturally with your cloned voice. --- 🚀 What this workflow does This automation links Veed.io’s video-generation API with multiple AI tools: Analyzes TikTok trends via Perplexity AI Writes a 10-second viral script using GPT-4 Generates your voiceover via ElevenLabs Uses Veed.io (Fabric 1.0 via FAL.ai) to animate your avatar and sync the lips to the voice Creates an engaging caption + hashtags for TikTok virality Publishes the video automatically via Blotato TikTok API Logs all results to Google Sheets for tracking --- 🧩 Setup Telegram Bot Create your bot via @BotFather Configure it as the trigger for sending your photo and theme Connect Veed.io Create an account on Veed.io Get your FAL.ai API key (Veed Fabric 1.0 model) Use HTTPS image/audio URLs compatible with Veed Fabric Other APIs Add Perplexity, ElevenLabs, and Blotato TikTok keys Connect your Google Sheet for logging results --- 🛠️ How to customize this workflow Change your Avatar: Upload a new image through Telegram, and Veed.io will generate a new talking version automatically. Modify the Script Style: Adjust the GPT prompt for tone (educational, funny, storytelling). Adjust Voice Tone: Tweak ElevenLabs stability and similarity settings. Expand Platforms: Add Instagram, YouTube Shorts, or X (Twitter) posting nodes. Track Performance: Customize your Google Sheet to measure your most successful Veed.io-based videos. --- 🧠 Expected Outcome In just a few seconds after sending your photo and theme, this workflow — powered by Veed.io — creates a fully automated TikTok video featuring your AI avatar with natural lip-sync and voice. The result is a continuous stream of viral short videos, made without cameras, editing, or effort. --- ✅ Import the JSON file in n8n, add your API keys (including Veed.io via FAL.ai), and start generating viral TikTok videos starring your AI avatar today! 🎥 Watch This Tutorial --- 📄 Documentation: Notion Guide Need help customizing? Contact me for consulting and support : Linkedin / Youtube

Dr. FirasBy Dr. Firas
39510

Generate song lyrics and music from text prompts using OpenAI and Fal.ai Minimax

Spark your creativity instantly in any chat—turn a simple prompt like "heartbreak ballad" into original, full-length lyrics and a professional AI-generated music track, all without leaving your conversation. 📋 What This Template Does This chat-triggered workflow harnesses AI to generate detailed, genre-matched song lyrics (at least 600 characters) from user messages, then queues them for music synthesis via Fal.ai's minimax-music model. It polls asynchronously until the track is ready, delivering lyrics and audio URL back in chat. Crafts original, structured lyrics with verses, choruses, and bridges using OpenAI Submits to Fal.ai for melody, instrumentation, and vocals aligned to the style Handles long-running generations with smart looping and status checks Returns complete song package (lyrics + audio link) for seamless sharing 🔧 Prerequisites n8n account (self-hosted or cloud with chat integration enabled) OpenAI account with API access for GPT models Fal.ai account for AI music generation 🔑 Required Credentials OpenAI API Setup Go to platform.openai.com → API keys (sidebar) Click "Create new secret key" → Name it (e.g., "n8n Songwriter") Copy the key and add to n8n as "OpenAI API" credential type Test by sending a simple chat completion request Fal.ai HTTP Header Auth Setup Sign up at fal.ai → Dashboard → API Keys Generate a new API key → Copy it In n8n, create "HTTP Header Auth" credential: Name="Fal.ai", Header Name="Authorization", Header Value="Key [Your API Key]" Test with a simple GET to their queue endpoint (e.g., /status) ⚙️ Configuration Steps Import the workflow JSON into your n8n instance Assign OpenAI API credentials to the "OpenAI Chat Model" node Assign Fal.ai HTTP Header Auth to the "Generate Music Track", "Check Generation Status", and "Fetch Final Result" nodes Activate the workflow—chat trigger will appear in your n8n chat interface Test by messaging: "Create an upbeat pop song about road trips" 🎯 Use Cases Content Creators: YouTubers generating custom jingles for videos on the fly, streamlining production from idea to audio export Educators: Music teachers using chat prompts to create era-specific folk tunes for classroom discussions, fostering interactive learning Gift Personalization: Friends crafting anniversary R&B tracks from shared memories via quick chats, delivering emotional audio surprises Artist Brainstorming: Songwriters prototyping hip-hop beats in real-time during sessions, accelerating collaboration and iteration ⚠️ Troubleshooting Invalid JSON from AI Agent: Ensure the system prompt stresses valid JSON; test the agent standalone with a sample query Music Generation Fails (401/403): Verify Fal.ai API key has minimax-music access; check usage quotas in dashboard Status Polling Loops Indefinitely: Bump wait time to 45-60s for complex tracks; inspect fal.ai queue logs for bottlenecks Lyrics Under 600 Characters: Tweak agent prompt to enforce fuller structures like [V1][C][V2][B][C]; verify output length in executions

Daniel NkenchoBy Daniel Nkencho
601

Automate invoice processing with OCR, GPT-4 & Salesforce opportunity creation

PDF Invoice Extractor (AI) End-to-end pipeline: Watch Drive ➜ Download PDF ➜ OCR text ➜ AI normalize to JSON ➜ Upsert Buyer (Account) ➜ Create Opportunity ➜ Map Products ➜ Create OLI via Composite API ➜ Archive to OneDrive. --- Node by node (what it does & key setup) 1) Google Drive Trigger Purpose: Fire when a new file appears in a specific Google Drive folder. Key settings: Event: fileCreated Folder ID: google drive folder id Polling: everyMinute Creds: googleDriveOAuth2Api Output: Metadata { id, name, ... } for the new file. --- 2) Download File From Google Purpose: Get the file binary for processing and archiving. Key settings: Operation: download File ID: ={{ $json.id }} Creds: googleDriveOAuth2Api Output: Binary (default key: data) and original metadata. --- 3) Extract from File Purpose: Extract text from PDF (OCR as needed) for AI parsing. Key settings: Operation: pdf OCR: enable for scanned PDFs (in options) Output: JSON with OCR text at {{ $json.text }}. --- 4) Message a model (AI JSON Extractor) Purpose: Convert OCR text into strict normalized JSON array (invoice schema). Key settings: Node: @n8n/n8n-nodes-langchain.openAi Model: gpt-4.1 (or gpt-4.1-mini) Message role: system (the strict prompt; references {{ $json.text }}) jsonOutput: true Creds: openAiApi Output (per item): $.message.content → the parsed JSON (ensure it’s an array). --- 5) Create or update an account (Salesforce) Purpose: Upsert Buyer as Account using an external ID. Key settings: Resource: account Operation: upsert External Id Field: taxid_c External Id Value: ={{ $json.message.content.buyer.tax_id }} Name: ={{ $json.message.content.buyer.name }} Creds: salesforceOAuth2Api Output: Account record (captures Id) for downstream Opportunity. --- 6) Create an opportunity (Salesforce) Purpose: Create Opportunity linked to the Buyer (Account). Key settings: Resource: opportunity Name: ={{ $('Message a model').item.json.message.content.invoice.code }} Close Date: ={{ $('Message a model').item.json.message.content.invoice.issue_date }} Stage: Closed Won Amount: ={{ $('Message a model').item.json.message.content.summary.grand_total }} AccountId: ={{ $json.id }} (from Upsert Account output) Creds: salesforceOAuth2Api Output: Opportunity Id for OLI creation. --- 7) Build SOQL (Code / JS) Purpose: Collect unique product codes from AI JSON and build a SOQL query for PricebookEntry by Pricebook2Id. Key settings: pricebook2Id (hardcoded in script): e.g., 01sxxxxxxxxxxxxxxx Source lines: $('Message a model').first().json.message.content.products Output: { soql, codes } --- 8) Query PricebookEntries (Salesforce) Purpose: Fetch PricebookEntry.Id for each Product2.ProductCode. Key settings: Resource: search Query: ={{ $json.soql }} Creds: salesforceOAuth2Api Output: Items with Id, Product2.ProductCode (used for mapping). --- 9) Code in JavaScript (Build OLI payloads) Purpose: Join lines with PBE results and Opportunity Id ➜ build OpportunityLineItem payloads. Inputs: OpportunityId: ={{ $('Create an opportunity').first().json.id }} Lines: ={{ $('Message a model').first().json.message.content.products }} PBE rows: from previous node items Output: { body: { allOrNone:false, records:[{ OpportunityLineItem... }] } } Notes: Converts discount_total ➜ per-unit if needed (currently commented for standard pricing). Throws on missing PBE mapping or empty lines. --- 10) Create Opportunity Line Items (HTTP Request) Purpose: Bulk create OLIs via Salesforce Composite API. Key settings: Method: POST URL: https://<your-instance>.my.salesforce.com/services/data/v65.0/composite/sobjects Auth: salesforceOAuth2Api (predefined credential) Body (JSON): ={{ $json.body }} Output: Composite API results (per-record statuses). --- 11) Update File to One Drive Purpose: Archive the original PDF in OneDrive. Key settings: Operation: upload File Name: ={{ $json.name }} Parent Folder ID: onedrive folder id Binary Data: true (from the Download node) Creds: microsoftOneDriveOAuth2Api Output: Uploaded file metadata. --- Data flow (wiring) Google Drive Trigger → Download File From Google Download File From Google → Extract from File → Update File to One Drive Extract from File → Message a model Message a model → Create or update an account Create or update an account → Create an opportunity Create an opportunity → Build SOQL Build SOQL → Query PricebookEntries Query PricebookEntries → Code in JavaScript Code in JavaScript → Create Opportunity Line Items --- Quick setup checklist 🔐 Credentials: Connect Google Drive, OneDrive, Salesforce, OpenAI. 📂 IDs: Drive Folder ID (watch) OneDrive Parent Folder ID (archive) Salesforce Pricebook2Id (in the JS SOQL builder) 🧠 AI Prompt: Use the strict system prompt; jsonOutput = true. 🧾 Field mappings: Buyer tax id/name → Account upsert fields Invoice code/date/amount → Opportunity fields Product name must equal your Product2.ProductCode in SF. ✅ Test: Drop a sample PDF → verify: AI returns array JSON only Account/Opportunity created OLI records created PDF archived to OneDrive --- Notes & best practices If PDFs are scans, enable OCR in Extract from File. If AI returns non-JSON, keep “Return only a JSON array” as the last line of the prompt and keep jsonOutput enabled. Consider adding validation on parsing.warnings to gate Salesforce writes. For discounts/taxes in OLI: Standard OLI fields don’t support per-line discount amounts directly; model them in UnitPrice or custom fields. Replace the Composite API URL with your org’s domain or use the Salesforce node’s Bulk Upsert for simplicity.

Le NguyenBy Le Nguyen
942