Structured Output Parser Workflows
233 Ergebnisse — ausschließlich quellenverknüpfte n8n-Referenzen
Confirm and log WhatsApp restaurant orders with Groq and Google Sheets
Quick Overview This workflow triggers on incoming WhatsApp orders, uses Groq to extract items and quantities, validates them against a live Google Sheets menu, logs the result to an orders sheet, and replies to the customer with either a clarification request or a priced confirmation (with optional VIP owner alerts). How it works Triggers when a new WhatsApp message is received on your WhatsApp Business Cloud number. Sends the message text to Groq (Llama 3.3) to extract a structured list of ordered items and quantities. Reads the latest menu (item name, price, availability) from a Google Sheets “Menu” tab. Matches extracted items to the menu, checks availability, calculates line subtotals and the total price, and compiles any validation issues. If there are issues, replies to the customer on WhatsApp asking for a corrected order and appends a “needs_clarification” entry to the Google Sheets “Orders” tab. If the order is valid, appends a “confirmed” entry to the Google Sheets “Orders” tab and sends a WhatsApp confirmation with the item summary and total. If the confirmed total meets or exceeds the VIP threshold, sends a separate WhatsApp notification to the owner number. Setup Conne
Create TikTok carousel posts with Upload Post, Google Sheets, Gemini, GPT-4.1
Quick overview This workflow runs daily to pull TikTok post and profile analytics from Upload Post, updates a Google Sheets performance log, uses Google Gemini and OpenAI to generate a new carousel concept and caption, creates four images via fal.ai, publishes to TikTok, and updates a persistent “Agent Skill” sheet. How it works Runs on a daily schedule (or manually) and loads product, Upload Post, and Google Sheets configuration values. Reads the Google Sheets “Performance Log,” selects posts missing stats or posted recently, fetches TikTok post analytics from the Upload Post API, and writes updated metrics back to the sheet. Aggregates the refreshed history, fetches TikTok profile analytics from Upload Post, reads the “Agent Skill” sheet, and compiles a single context summary. Uses Google Gemini (via a structured JSON output) to generate a TikTok carousel plan including caption, strategy notes, four slide prompts, hashtags, and an engagement question. Sends the four prompts to fal.ai to generate four 9:16 PNG carousel images and collects their hosted image URLs. Uses OpenAI (GPT-4.1) to write a short TikTok photo-post description, publishes the carousel to TikTok via Upload Post
Create and approve AI social posts with OpenAI, Telegram and Blotato
Quick overview This workflow runs daily to generate an AI-written social post and DALL·E image with OpenAI, sends it to Telegram for approval, and—if approved—publishes it to LinkedIn, X, Instagram, Facebook, and Threads via Blotato while logging outcomes to Google Sheets. How it works Runs every 24 hours on a schedule trigger. Uses OpenAI to generate a single content topic (title, angle, and hook) based on your niche, audience, and language settings. Uses OpenAI to write one social post with a separate hashtag line and an image prompt in your specified tone and style. Uses OpenAI Images (DALL·E) to generate a matching image and extracts a shareable image URL. Sends the post text, hashtags, and image link to Telegram and waits for you to approve or skip within the configured time window. If approved, publishes the post with the image to LinkedIn, X, Instagram, Facebook, and Threads using Blotato and appends a “Published” log row to Google Sheets. If skipped or not approved, appends a “Skipped” log row to Google Sheets. Setup Add OpenAI API credentials and ensure the selected chat model and image model in the configuration match models available to your account. Add Telegram credent
Generate CCTV-style animal TikTok videos with OpenAI, Seedance and Blotato
Quick Overview This workflow runs every four hours to generate a CCTV-style animal video concept with OpenAI, renders an 8-second vertical clip using AtlasCloud Seedance, then publishes it to TikTok via Blotato and sends a preview to Telegram. How it works Runs every 4 hours on a schedule. Uses OpenAI to create a hyper-realistic CCTV-style image and video prompt based on a random (or overridden) animal, location, action, and a current timestamp overlay. Starts a Seedance text-to-video render on AtlasCloud with the generated prompt, duration, aspect ratio, and optional audio. Polls AtlasCloud at a fixed interval until the render status is completed or succeeded and retrieves the output video URL. Uses OpenAI to generate a TikTok-ready title and exactly five hashtags from the video concept. Publishes the generated video to TikTok via Blotato with the title/hashtags and the selected privacy level, flagged as AI-generated. Sends the rendered video and title as a preview to a specified Telegram chat. Setup Add OpenAI API credentials and set the prompt and caption models in the Configuration values if you want to change them. Add an AtlasCloud API key (HTTP Header Auth) and ensure it is
Route WhatsApp support requests with Claude, Supabase, HubSpot and Slack
Quick overview This workflow turns incoming WhatsApp messages (text, audio, images, and documents) into actionable support interactions by extracting content, detecting sentiment, and using an AI agent with a Supabase knowledge base to route requests to Google Calendar, HubSpot, Slack, Gmail, and back to WhatsApp. How it works Receives incoming WhatsApp messages via a WhatsApp webhook trigger. Routes the message by type and converts it into text using OpenAI Whisper for audio, OpenAI GPT-4o Vision for images, or a PDF extractor for documents (or passes text messages through). Loads the sender’s session context from Supabase and runs sentiment/language detection on the extracted text. Uses a LangChain AI agent (Claude) with Supabase vector search (with OpenAI embeddings and Cohere reranking) plus tool access (Google Calendar, HubSpot, Slack, and Gmail) to produce a structured intent, response, and metadata. Logs conversation analytics (intent, sentiment, message type, and response length) to an n8n Data Table. Routes by intent to either create a Google Calendar event, create/update a HubSpot contact, email a document via Gmail (and upload it to Google Drive), create a HubSpot ticket
Send prescription PDFs to patients via WhatsApp with Google Gemini and Supabase
Quick overview This workflow listens for incoming WhatsApp messages from clinic staff, uses Google Gemini to extract prescription details, generates a prescription PDF via an HTML-to-PDF API, uploads it to a public URL, sends the document back over WhatsApp, and logs outcomes to Supabase. How it works Triggers whenever a new WhatsApp message is received on your WhatsApp Business Cloud number. Captures configuration values and the incoming message text and sender number. Uses Google Gemini to extract patient name, doctor name, and medication details into structured data and flags any missing required fields. If required details are missing, replies to staff on WhatsApp with what to fix and inserts a needs_review record into Supabase. If details are complete, renders a prescription HTML document, converts it to a PDF via an external HTML-to-PDF API, and uploads the PDF to an external file-hosting API. Sends the prescription PDF to the sender on WhatsApp and inserts a “sent” log record (including the PDF URL) into Supabase. Setup Connect WhatsApp Business Cloud credentials for the WhatsApp trigger and the WhatsApp send steps, and configure the workflow’s webhook in Meta/WhatsApp so in
Log email receipts to Google Sheets with Gmail and OpenAI
Quick Overview This workflow logs Gmail receipt attachments to Google Sheets by extracting structured fields with OpenAI (including category selection from your list), prevents duplicates by checking the ledger, and sends a monthly spending summary email via Gmail based on last month’s rows. How it works Triggers when a Gmail message matching has:attachment subject:receipt arrives or when a schedule runs on the 1st of the month at 8am. For receipt emails, looks up the Gmail message ID in Google Sheets and stops if the receipt is already logged. If the receipt is new, extracts text from PDF attachments or transcribes image receipts using OpenAI vision. Sends the receipt text to OpenAI to extract receipt_date, merchant, amount, currency, and a category chosen only from your predefined category list. Appends one cleaned expense row to Google Sheets with the extracted fields plus a processed timestamp, Gmail message ID, and the original email subject. For the monthly run, reads the full Google Sheets ledger, totals last month’s expenses by category, and emails the summary to you via Gmail. Setup Connect Gmail credentials for the Gmail Trigger and for sending the monthly summary email.
Validate startup ideas and generate business plans from Telegram with Perplexity and GPT-4o
Quick overview This workflow receives startup ideas via a Telegram bot, uses OpenAI (GPT-4o and GPT-4o mini) plus Perplexity to score, research, and validate them, then generates a full business plan in Google Docs, logs key results to Google Sheets, and returns the plan link in Telegram. How it works Receives a Telegram message containing a raw startup idea. Uses OpenAI (GPT-4o mini) to structure the idea, detect its language, and assign an initial 1–10 idea score with improvement feedback. If the idea score is below the configured threshold, sends the score reasoning back to Telegram and stops. If the score meets the threshold, asks in Telegram for approval to run full research and plan generation. After approval, uses Perplexity to research the target market and competitors and then uses OpenAI (GPT-4o) to structure the findings, score market and competition, and generate differentiation and moat notes. Uses OpenAI (GPT-4o mini) to validate demand, risks, and go-to-market approach, then computes a composite score and recommendation. Uses OpenAI (GPT-4o) to write a full business plan, saves it to a Google Doc, appends a summary row to Google Sheets, and sends the Google Doc link
Translate YouTube video metadata into multiple languages with OpenAI
Quick overview This workflow fetches a YouTube video’s title and description, uses OpenAI to translate them into multiple target languages, and updates the video’s YouTube localizations via the YouTube Data API, returning a YouTube Studio link to review the translations. How it works Runs manually when you execute the workflow. Reads the configured YouTube video ID, default language, target languages, and OpenAI model. Fetches the video’s current metadata (including existing localizations) from YouTube. Removes the default language from the target language list and prepares the language-code keys required for the translation output. Sends the title and description to OpenAI and parses the response as structured JSON with a translated title and description per language code. Updates the video on YouTube by writing the translated metadata into the video’s localizations via a single YouTube API request. Outputs a YouTube Studio translations URL so you can review the updated localizations. Setup Add YouTube OAuth2 credentials with permission to edit videos on the target channel. Add an OpenAI API key credential and choose a model name in the configuration. Set the video_id, default_lan
Qualify WhatsApp real estate leads with Gemini and Supabase
Quick overview This workflow triggers on incoming WhatsApp messages, uses Google Gemini to extract and score real estate lead details, stores each lead in Supabase, and sends a WhatsApp alert to the agent only for high-intent inquiries based on configurable score thresholds. How it works Triggers when a new WhatsApp message is received via the WhatsApp Business Cloud API. Pulls the sender name, phone number, and message text and applies your configured scoring thresholds and agent contact number. Sends the message to Google Gemini to extract structured lead details (budget range, property type, location preference, timeline, and intent score). Inserts the extracted lead data into a Supabase leads table for tracking. Routes the lead based on the intent score and sends a formatted WhatsApp message to the agent for high-intent leads. Leaves medium- and low-intent leads logged in Supabase without sending an agent alert. Setup Connect WhatsApp Business Cloud API credentials for both the incoming-message trigger and the WhatsApp send action, and activate the workflow to generate the webhook for your WhatsApp app. Add a Google Gemini (PaLM) API credential for the Gemini chat model. Add a
CDHandle LinkedIn inbox RAG replies with Unipile, Claude, Pinecone, and Airtable
Quick overview Automatically ingest your knowledge base, monitor your LinkedIn company page inbox, and reply to leads and customers with AI-drafted, source-grounded answers. Classifies every message, checks for spam, drafts replies using Claude and Pinecone RAG, and lets you auto-send or review each reply in Airtable. How it works A Google Drive trigger watches a folder for new or updated documents, downloading each file automatically as it's added. Each document is split into chunks, embedded, and upserted into a Pinecone vector index, building a searchable company knowledge base. A webhook receives every new LinkedIn message sent to your company page via Unipile, filtering out messages you sent yourself. The last few messages in the conversation are fetched and formatted into a chronological transcript for context. An AI agent classifies the message's intent (lead, support, partnership, recruiting, spam, or other) using shared conversation memory. A second AI agent drafts a reply, searching the Pinecone knowledge base for relevant facts before answering, and saves the draft to Airtable. Depending on a configurable auto-send toggle, the reply is sent immediately via Unipile, or he
Triage support tickets with OpenRouter (OpenAI o3-mini) and ClickUp
Quick overview This workflow collects support tickets via an n8n form, uses OpenRouter (OpenAI o3-mini) to assess priority, severity, impact, and suggested diagnostics, and then creates a formatted task in ClickUp with the AI-generated incident details. How it works Receives a new ticket submission through an n8n form with subject, description, company, name, and email. Normalizes the submitted fields into a consistent ticket payload (customer name/email, subject, and description). Validates that the ticket has a description and that the email address contains an “@” symbol. Sends the ticket subject and description to OpenRouter (OpenAI o3-mini) to generate incident metadata, priority (1–4), impact, severity, affected service, and recommended diagnostics and next steps in structured JSON. Creates a new task in ClickUp using the ticket subject as the task name and adding the customer details plus the AI assessment and recommendations to the task description. Setup Create or select an n8n Form Trigger and share the form URL with users who should be able to submit tickets (configure any access restrictions as needed). Add an OpenRouter credential to the OpenRouter chat model node and
Create AI music videos from a prompt with OpenAI, Kie.ai and BeatAPI
Quick overview This workflow collects a music video theme via an n8n form, uses OpenAI to generate music, image, and video prompts, then calls Kie to create an instrumental track and a Nano Banana image, and finally uses BeatAPI to render a publish-ready AI music video. How it works Receives a submission from an n8n Form Trigger with inputs like theme, genre, mood, duration, aspect ratio, and quality. Uses OpenAI (via an AI Agent with a structured output schema) to generate a Suno-style music prompt, an image prompt, a BeatAPI music-video prompt, and YouTube metadata. Sends the generated music prompt and settings to Kie to start audio generation, then polls Kie until the track is ready and validates the duration is within 10–180 seconds. Sends the generated image prompt to Kie (Nano Banana model) to create key art, then polls Kie until an image URL is returned. Creates a BeatAPI AI music video task using the generated audio URL, one or two image URLs (generated image plus optional cover image), and the BeatAPI prompt, then polls BeatAPI until the render succeeds. Returns the final video URL along with YouTube-ready title, description, tags, and provider task IDs in a single publish
Send daily AI-personalized cold emails with Google Places and OpenAI
Quick overview This workflow runs every morning to find local businesses via Google Places, pulls their website content to extract an email address, uses OpenAI to draft a personalised automation-focused outreach message, then sends it via SMTP while tracking contacts in an n8n Data Table. How it works Runs daily at 9:00 and sets the campaign parameters (rotating niche and city, sender details, daily send cap, and delay between emails). Searches Google Places for businesses matching the selected niche and location, then iterates through each returned place. Keeps only businesses with a website, fetches the website HTML, and extracts a likely contact email address plus a short text snippet. Filters out leads without an email and skips any email address already logged in the n8n Data Table. Limits the run to the configured daily cap and processes leads in batches. Uses OpenAI to generate a structured subject and body for a personalised cold email based on the business snippet and your campaign details. Sends the email via SMTP with your signature and unsubscribe note, logs the contact in the Data Table, and waits the configured delay before sending the next email. Setup Add credentia
WSummarize Telegram-shared webpages with OpenAI GPT-4o
Quick overview This workflow listens for Telegram messages containing a URL, fetches and cleans the webpage content, summarizes it with an OpenAI chat model, and replies in the same Telegram chat with a formatted summary including key points and action items. How it works Triggers when a new Telegram message is received by your bot. Extracts the first http/https URL from the message text and captures the sender’s chat ID. If no valid URL is found, sends an error message back to the same Telegram chat. If a valid URL is found, fetches the webpage HTML and extracts the title and cleaned body text while removing common boilerplate elements. Truncates the extracted text to a configurable maximum length and sends the title, URL, and text to OpenAI to generate a structured JSON summary. Formats the JSON into Telegram HTML (title, source link, summary, key points, and insights/action items) and sends it back to the original chat. Setup Create a Telegram bot and add a Telegram API credential in n8n, then select it in both the Telegram Trigger and the send-message steps. Add an OpenAI API credential in n8n and select it for the OpenAI chat model used for summarization. Activate the workflow
Triage Gmail and draft replies with Ollama, Telegram and Google Sheets
Quick Overview This workflow pulls unread Gmail messages, uses a local Ollama Llama 3.1 model to classify and extract action items, drafts a reply, and sends an approval request to Telegram before creating a threaded Gmail draft and logging the decision to Google Sheets. How it works Runs manually and fetches up to 10 unread emails from Gmail. Normalizes each email into a consistent structure (sender, subject, body, message ID, and thread ID) and classifies it with Ollama into urgent, action_needed, fyi, or spam. Continues only for emails classified as urgent or action_needed and marks all other categories as skipped. For actionable emails, uses Ollama to extract a task title and deadline and to draft a short professional reply as structured JSON. Sends the classification, extracted task/deadline, and draft reply to Telegram and waits for you to approve or decline. If approved, creates a Gmail draft reply in the original thread, adds a Gmail label (STARRED), and logs the approval/rejection plus details to Google Sheets. Setup Connect your Gmail OAuth2 credential and ensure it has permission to read messages, create drafts, and add labels. Install and run Ollama, pull the llama3.1:l
Triage support tickets with Claude Sonnet, Infini Analytics, and Slack
Quick overview Automatically triage incoming support tickets with an AI agent powered by Claude Sonnet. High-confidence responses are sent instantly, low-confidence cases escalate to a human on Slack, and every step is logged in Infini Analytics for full observability. How it works A webhook receives the incoming support ticket and generates a unique execution_id to correlate every event of this run. Infini Analytics logs a START event for the execution. An AI Agent (Claude Sonnet) reads the ticket, optionally queries a knowledge base data table, and drafts a response with a confidence score. The agent logs structured ticket data via a data table tool (Infini Analytics records this as an EVENT). An IF node checks the confidence: high-confidence tickets get an automatic reply and log an EVENT; low-confidence tickets escalate to a human on Slack and log a WARNING. Infini Analytics logs the final END event, or an ERROR event if any step failed along the way. Setup Import this workflow into your n8n instance and open it in the editor. Create an Anthropic credential with your Claude API key and select it in the AI Agent node Create an Infini Analytics credential with your Bearer Token,
Schedule YouTube Shorts to Pinterest with Apify, Gemini, and an OpenAI model
Quick overview This workflow receives YouTube Shorts webhook events, downloads the video via Apify, uses Google Gemini and an OpenAI-compatible chat model to generate Pinterest SEO metadata and choose a board, then schedules a Pinterest video pin through Zernio (Late) for publishing. How it works Receives a webhook request for YouTube PubSubHubBub verification and responds with the provided hub challenge. Receives a POST webhook event for a new YouTube video and extracts the title, video ID, channel ID, and video URL from the feed payload. Uses Apify to fetch and process the YouTube Short and returns a downloadable MP4 URL. Sends the video to the Google Generative Language (Gemini) API to generate an analysis used for Pinterest-focused metadata. Uses a LangChain LLM chain (with structured output validation) to produce a Pinterest-ready title, description, hashtags, and a board selection from a fixed list. Maps the selected board name to a Pinterest board ID and schedules the video post to Pinterest via the Zernio (Late) API. Setup Create the YouTube webhook/PubSubHubbub subscription and configure it to call this workflow’s webhook URL at the manga-monarch-webhook path for both veri
Create AI image and video posts from Telegram with GPT-4o, AtlasCloud and Blotato
Quick overview This workflow receives a post idea via Telegram, uses OpenAI (GPT-4o mini) to generate platform-specific copy plus image/video prompts, creates media through AtlasCloud (GPT Image 2 and Seedance 2.0), requests approvals in Telegram, and publishes the approved content to seven social platforms via Blotato. How it works Triggers when a new Telegram message arrives with a post description. Uses OpenAI (GPT-4o mini) to generate copy for X, LinkedIn, Facebook, Threads, Instagram, TikTok, and YouTube Shorts, plus an English image prompt and video prompt. Sends the prompts to AtlasCloud to generate a square image (GPT Image 2) and a vertical short video (Seedance 2.0), polling each prediction until it completes. Sends the generated image and video back to Telegram for preview and asks for approval, automatically regenerating the declined media. Publishes the approved image to X, LinkedIn, Facebook, and Threads via Blotato and confirms completion in Telegram. Publishes the approved video to Instagram Reels, TikTok, and YouTube Shorts via Blotato (with synthetic/AI flags enabled where configured) and confirms completion in Telegram. Setup Create a Telegram bot with BotFather,
Draft Google review reply emails with Gmail, OpenAI, and Google Sheets
Quick Overview This workflow monitors Gmail for Google Business Profile “new review” alert emails, uses OpenAI to extract the rating and review text and draft a policy-safe reply, then emails the draft to you and logs the review and reply draft to Google Sheets. How it works Triggers when a new Google Business Profile review notification email arrives in Gmail. Looks up the email’s message ID in a Google Sheets ledger to prevent drafting duplicate replies. Sends the email content and your configured business details to OpenAI to extract the rating, reviewer first name, and review text and generate a reply draft as structured JSON. Emails you a formatted message containing the review excerpt, the suggested reply, and a link to your Google reviews page. Appends a new row to Google Sheets with the message ID, extracted review details, the reply draft, a timestamp, and a “draft-sent-to-owner” status. Setup Connect your Gmail account and ensure Google review alerts are delivered from [email protected]. Add an OpenAI API key/credential and select the chat model in the OpenAI node. Create a Google Sheets “reply ledger” with columns like processed_at, gmail_message_id, rat
Vet community membership applications with Google Gemini, Telegram, and Sheets
Quick overview This workflow collects community membership applications via an n8n Form, vets each candidate with Google Gemini using Tavily web search and an Apify LinkedIn lookup, requests human approval in Telegram when needed, sends invite or rejection emails via Gmail, and logs outcomes to Google Sheets. How it works Receives a submission from an n8n Form that captures applicant details such as name, email, company, role, birthday, and additional notes. Normalizes the submitted fields and passes the candidate profile to a Google Gemini-powered AI Agent that can use Tavily web search, an Apify LinkedIn search actor, a calculator, and short-term memory. Produces a structured vetting result (including summary, decision status, and a community-fit score) using a structured output schema. If the AI decision is “Immediate approval”, sends the candidate summary to Telegram and waits for a response before continuing. If the AI decision is not “Immediate approval”, sends the summary to Telegram and asks the reviewer to reply with 1 (approve), 2 (reject), or 3 (do nothing). Sends an approval/invite email or a rejection email via Gmail based on the Telegram response, or stops without act
Validate invoice tax compliance from Gmail using Groq, Notion, and Google Sheets
Quick overview This workflow monitors a Gmail inbox for PDF invoices, uses Groq (Llama 3.3) to extract structured invoice fields, validates the invoice against Purchase Orders in Google Sheets and tax rules in Notion, logs an audit entry to Notion, and notifies Slack and the supplier based on pass/fail. How it works Triggers every minute when a new email arrives in Gmail and downloads the attached PDF invoice. Extracts text from the PDF and sends it to Groq (Llama 3.3) to parse invoice fields into a structured JSON schema, including SEZ/LUT detection. Looks up the referenced PO number in Google Sheets and fetches applicable tax rules from a Notion database. Validates the invoice against the PO (existence, approval status, currency, and total amount) and against Notion tax rules (SEZ/import/domestic matching and GST expectations). Writes the compliance result, matched rule, and any violations to a Notion “Compliance Audit Log” database. If the invoice passes, posts an approval message to Slack and archives the PDF to a Google Drive folder. If the invoice fails, posts a blocking alert to Slack and emails the supplier a rejection notice listing the violations. Setup Connect your Gmail
Predict supplier churn and generate retention plans with Notion, Slack, and Groq
Quick overview This workflow runs weekly to pull supplier activity from Notion, engineer churn-risk features, and use Groq-hosted LLM prompts to classify supplier churn risk and generate retention plans. It updates each supplier record in Notion and posts a Slack alert only when a supplier is rated High risk. How it works Runs every week on a schedule. Retrieves all supplier records from a Notion database. Calculates behavioral risk features such as PO trend, engagement score, and delivery/financial/contract-risk flags from the Notion fields. Processes suppliers one at a time and sends the features to a Groq LLM to return structured JSON with churn risk level, risk score, confidence, and key drivers. Sends the churn assessment to a second Groq LLM prompt to generate a structured retention plan with priority, recommended actions, suggested owner, and summary. If the supplier’s churn risk is High, posts a formatted alert with drivers and recommended actions to a Slack channel. Updates the supplier’s Notion page with the churn risk, risk score, key drivers, retention plan, and a “Last Evaluated” timestamp, then continues to the next supplier. Setup Connect your Notion credentials and
Review contract risk and rewrite clauses with Groq, Google Drive, Notion, and Slack
Quick overview This workflow watches a Google Drive folder for newly uploaded contract PDFs, extracts key clauses with Groq-hosted LLMs, compares them to standard clauses stored in Notion, generates a structured risk report and clause rewrites, sends a Slack alert for high-risk contracts, and saves the review to Notion. How it works Triggers when a new file is created in a specific Google Drive folder. Downloads the file and extracts text from it (using PDF text extraction when the uploaded file is a PDF). Uses a Groq-hosted LLM to extract key contract clauses (payment terms, penalties, SLA, termination, renewal, liability, and data protection) into structured JSON. Pulls the company’s standard clause library from Notion and compares each extracted vendor clause to the corresponding standard text. Uses a Groq-hosted LLM to generate a structured legal risk report with an overall risk level, executive summary, clause-by-clause risk review, missing protections, and negotiation recommendations. Rewrites medium- and high-risk clauses with a Groq-hosted LLM to produce negotiation-ready alternative wording aligned to the Notion standards. If the overall risk level is High, posts a formatt