Workflow Templates

Automatically Scrape Make.com Job Board with GPT-5-mini Summaries & Email Digest Overview Who is this for? Make.com consultants, automation specialists, and freelancers who want to catch new client opportunities without manually checking the forum. What problem does it solve? Scrolling through forum posts to find jobs wastes time. This automation finds new postings, uses AI to summarize what clients need, and emails you a clean digest. How it works: Runs on schedule → scrapes the Make.com professional services forum → filters jobs from last 7 days → AI summarizes each posting → sends formatted email digest. Use Cases Freelancers: Get daily job alerts without forum browsing, respond to opportunities faster Agencies: Keep sales teams informed of potential clients needing Make.com expertise Job Seekers: Track contract and full-time positions requiring Make.com skills Detailed Workflow Scraping: HTTP module pulls HTML from the Make.com forum job board Parsing: Extracts job titles, dates, authors, and thread links Filtering: Only jobs posted within last 7 days pass through (configurable) AI Processing: GPT-5-mini analyzes each post to extract: Project type Key requirements Complexity level Budget/timeline (if mentioned) Email Generation: Aggregates summaries into organized HTML email with direct links Delivery: Sends via SMTP to your inbox Setup Steps Time: ~10 minutes Requirements: OpenRouter API key (get one here) SMTP credentials (Gmail, SendGrid, etc.) Steps: Import template Add OpenRouter API key in "OpenRouter Chat Model" node Configure SMTP settings in "Send email" node Update recipient email address Set schedule (recommended: daily at 8 AM) Run test to verify Customization Tips Change date range: Modify filter from 7 days to X days: {{now - X days}} Keyword filtering: Add filter module to only show jobs mentioning "API", "Shopify", etc. AI detail level: Edit prompt for shorter/longer summaries Multiple recipients: Add comma-separated emails in Send Email node Different AI model: Switch to Gemini or Claude in OpenRouter settings Team notifications: Add Slack/Discord webhook instead of email

Quick Overview This workflow listens for emergency equipment breakdown requests via Telegram, emails multiple rental vendors through Gmail to request quotes, uses Azure OpenAI to compare responses, and sends the project manager an approval email with Approve/Reject links before notifying the vendor and foreman. How it works Triggers when a foreman sends a /breakdown command to a Telegram bot. Parses and normalizes the request details (machine type, site, duration, urgency) and generates a unique breakdown reference. Sends a quote-request email to each vendor via Gmail and waits for the vendor reply window to close. Aggregates collected vendor quotes, filters out unavailable options, and sends the shortlist to Azure OpenAI to produce a ranked recommendation and HTML comparison table. Emails the project manager via Gmail with the comparison and one-click approval and rejection links, then waits for the response. If approved, emails a booking confirmation to the recommended vendor and sends a Telegram confirmation message to the foreman; if rejected, emails the project manager a manual-review notice. If the workflow errors at any point, posts an error alert to a designated Slack channel. Setup Add Telegram Bot credentials for the Telegram trigger and ensure the /breakdown command is configured for your bot. Add Gmail OAuth2 credentials and update all Gmail nodes with the correct sender account and real recipient addresses (vendors and project manager). Add Azure OpenAI credentials (endpoint, deployment/model, and API key) for the quote analysis step. Replace the placeholder vendor names and email addresses in the vendor list with your actual suppliers and ensure vendor replies are captured and mapped into the workflow’s quote aggregation. Replace the placeholder approval webhook base URL (YOUR-N8N-INSTANCE) with your live n8n domain so the Approve/Reject links route back to your instance. Add Slack OAuth2 credentials and set the target channel for workflow error notifications.

LinkedIn Inbox Triage (Gmail Label to Notion + Slack) This n8n template demonstrates how to use AI to triage LinkedIn emails in your Gmail inbox, so you only see the messages worth your time. It filters out automated noise, scores sales likelihood, drafts quick replies for real conversations, stores everything in Notion, and sends you a Slack DM for items you should answer quickly. Good to know This workflow sends email content to an LLM. Do not use it with sensitive mailboxes unless you are comfortable with that. Cost depends on your model choice and token usage. The body is currently limited to 4000 characters to control spend. If you want a shorter run window, adjust the receivedAfter filter. How it works Runs on a daily schedule. Pulls emails from Gmail using a label you define (example: LinkedIn). Applies two filters: Keeps only invitations and messages Removes common automated notifications Fetches the full email body for better classification. Sends the message to an AI agent that returns strict structured JSON: action (reply_quick, review, ignore, block) relevancy_score (0 to 100) sales_likelihood (0 to 1) summary optional reply_draft Applies a quality gate to keep high signal messages. Writes the output to a Notion database as a ticket. Sends a Slack DM only for items marked reply_quick. How to use Create a Gmail label that captures LinkedIn emails, then add the label id to the Gmail node. Create a Notion database with fields matching the Notion node mapping. Connect your OpenAI, Gmail, Notion, and Slack credentials in n8n. Run once manually to verify mapping, then enable the workflow. Requirements Gmail account OpenAI API credentials (or compatible model node) Notion database Slack account Customising this workflow Make it more aggressive by increasing the sales threshold or raising the relevancy cutoff. Add more filter phrases for your own LinkedIn email language. Swap Slack DM for a channel post, or send a daily digest instead of per message. Add a redaction step before the AI node if you want to remove signatures or quoted replies.

Description Submit any podcast episode recording URL along with the episode number, podcast name, host name, episode topic, niche, and CTA link — and optionally a guest name and title if it is an interview episode — and the workflow transcribes and analyzes the full episode automatically. WayinVideo's Transcription API returns a speaker-labeled, timestamped transcript which GPT-4o-mini reads to write eight labeled show notes sections — episode title, SEO description, key takeaways, resources mentioned, timestamped highlights, guest bio, CTA block, and a fully compiled show notes document ready to paste. The GPT prompt adapts automatically for interview episodes versus solo episodes based on whether you provided a guest name. Everything is saved to Google Sheets as one row with Status set to Draft — ready to copy directly into Spotify for Podcasters, Apple Podcasts, or your episode page. Built for podcast producers, solo hosts, and content teams who spend 30–60 minutes per episode writing show notes manually. What This Workflow Does Transcribes the full episode with speaker labels and timestamps** — WayinVideo returns every spoken word with the speaker name and MM:SS timestamp so GPT can extract highlights and quotes from the actual episode content Detects interview vs solo format automatically** — If you provide a guest name in the form, the prompt adapts to write a guest bio and frame the show notes as an interview — if you leave it blank, it treats the episode as solo Generates a compiled FULL_SHOW_NOTES section** — All eight sections are assembled into one complete, copy-paste-ready show notes document — the full block is saved in its own column so you can paste it directly into Spotify, Apple, or a website without reassembling it manually Extracts timestamped highlights from the actual transcript** — GPT picks 6–8 of the most interesting moments from the episode using exact MM:SS timestamps from the transcript — not approximate or invented times Writes an SEO episode description with keyword in the first sentence** — The 150–200 word description follows podcast platform SEO best practice so your episode is discoverable in search Logs only real resources mentioned in the episode** — GPT is instructed to extract only resources, links, or tools that were explicitly named in the recording — no invented or generic resources Saves 19 fields per episode to Google Sheets** — All eight content sections plus episode metadata, word count, duration, recording URL, and CTA link are saved in one row per episode Setup Requirements Tools Needed n8n instance (self-hosted or cloud) WayinVideo account with API access OpenAI account with GPT-4o-mini API access Google Sheets (one sheet with a tab named Show Notes Library) Credentials Required WayinVideo API key (pasted into 2. WayinVideo — Submit Transcription and 4. WayinVideo — Get Transcript Results) OpenAI API key Google Sheets OAuth2 > ⚠️ WayinVideo API key appears in 2 steps — replace YOUR_WAYINVIDEO_API_KEY in both 2. WayinVideo — Submit Transcription and 4. WayinVideo — Get Transcript Results. Missing either one will cause the workflow to fail. Estimated Setup Time: 15–20 minutes Step-by-Step Setup Import the workflow — Open n8n → Workflows → Import from JSON → paste the workflow JSON → click Import Get your WayinVideo API key — Log in to your WayinVideo account → go to Account Settings → copy your API key Add your WayinVideo API key to node 2 — Open node 2. WayinVideo — Submit Transcription → find the Authorization header value Bearer YOUR_WAYINVIDEO_API_KEY → replace YOUR_WAYINVIDEO_API_KEY with your actual key Add your WayinVideo API key to node 4 — Open node 4. WayinVideo — Get Transcript Results → find the same Authorization header → replace YOUR_WAYINVIDEO_API_KEY with the same key Connect OpenAI — Open node 9. OpenAI — GPT-4o-mini Model → click the credential dropdown → add your OpenAI API key → test the connection Create your Google Sheet tab — Open your Google Sheet → add a tab named exactly Show Notes Library → add these 19 column headers in row 1: Episode Number, Episode Title, Podcast Name, Host, Guest, Topic, Duration, Recording URL, SEO Description, Key Takeaways, Resources Mentioned, Timestamped Highlights, Guest Bio, CTA Block, Full Show Notes, Word Count, CTA Link, Generated On, Status Get your Google Sheet ID — Open your Google Sheet in a browser → copy the string between /d/ and /edit in the URL Connect Google Sheets — Open node 11. Google Sheets — Save Show Notes Library → click the document field → replace YOUR_GOOGLE_SHEET_ID by selecting your spreadsheet or entering the Sheet ID manually → click the credential dropdown → add Google Sheets OAuth2 → authorize access Activate the workflow — Toggle the workflow to Active → copy the Form URL from node 1. Form — Episode Recording + Details → open it in a browser to submit your first episode How It Works (Step by Step) Step 1 — Form: Episode Recording + Details You open the form URL and fill in up to nine fields. Five are required for all episodes: the recording URL, episode number, podcast name, host name, episode main topic, podcast niche, and CTA or subscribe link. Two are optional: Guest Name and Guest Title / Company — leave these blank for solo episodes and fill them in for interview episodes. The workflow uses whether you provided a guest name to decide how to frame the show notes. Step 2 — HTTP: WayinVideo — Submit Transcription The episode recording URL is sent to WayinVideo's Transcription API using the /v2/transcripts endpoint. WayinVideo accepts the job and returns a task ID. The transcription returns speaker-labeled segments — each with a speaker name, start time, end time, and spoken text. Step 3 — Wait: 90 Seconds The workflow pauses 90 seconds before the first status check, giving WayinVideo time to transcribe the full episode. Longer episodes may need more polling cycles after this initial wait. Step 4 — HTTP: WayinVideo — Get Transcript Results A GET request checks the transcription results endpoint using the task ID from step 2. It returns the current status and, once complete, the full transcript array. Step 5 — IF: Transcription Complete? This is the polling gate. If the status equals SUCCEEDED (YES path), the transcript is ready and the workflow moves forward to formatting. If still processing (NO path), the workflow routes to 6. Wait — 30 Seconds Retry which pauses 30 seconds then loops back to step 4. This repeats until SUCCEEDED. Step 6 — Wait: 30 Seconds Retry When the transcript is not yet ready, the workflow waits 30 seconds then returns to step 4 for another check. Step 7 — Code: Format Transcript Each transcript segment is formatted as [Speaker | MM:SS] Spoken text and joined into a single readable block. Total episode duration is calculated and formatted as MM:SS. A word count of the transcript and unique speakers list are also extracted. An isInterviewEpisode flag is set to true if the Guest Name field was filled in — this flag is passed to the GPT prompt so it knows whether to write a guest bio or a solo note. All nine form inputs are packaged alongside the formatted transcript. Step 8 — AI Agent: Write Show Notes GPT-4o-mini receives the full formatted transcript as the main input and the episode context in the system prompt. The system prompt includes an inline conditional — if isInterviewEpisode is true, the guest name and title are shown; if false, it shows "Format: Solo Episode". GPT writes eight labeled sections: EPISODE_TITLE (60–70 characters with the main keyword), SEO_DESCRIPTION (150–200 words with keyword in the first sentence), KEY_TAKEAWAYS (5–8 specific actionable bullets), RESOURCES_MENTIONED (only things explicitly mentioned in the recording), TIMESTAMPED_HIGHLIGHTS (6–8 moments with exact MM:SS from the transcript), GUEST_BIO (2–3 sentences based only on what was said, or "Solo episode — no guest"), CTA_BLOCK (3–4 line subscribe and review prompt with the CTA link), and FULL_SHOW_NOTES (all sections compiled into one complete copy-paste block). Step 9 — OpenAI: GPT-4o-mini Model This is the language model powering the show notes generation. Step 10 — Code: Parse Show Notes Output All eight labeled sections are extracted from the GPT output using regex. If EPISODE_TITLE, SEO_DESCRIPTION, or FULL_SHOW_NOTES are missing, the step throws a clear error. Word count is calculated from the FULL_SHOW_NOTES section. All episode metadata from step 7 is also packaged for the sheet row. Step 11 — Google Sheets: Save Show Notes Library One row is appended to your Show Notes Library tab with all 19 columns: episode number, episode title, podcast name, host, guest, topic, duration, recording URL, SEO description, key takeaways, resources mentioned, timestamped highlights, guest bio, CTA block, full show notes, word count, CTA link, generation timestamp, and Status set to Draft. Key Features ✅ Adapts automatically for interview vs solo episodes — The workflow detects whether a guest name was provided and adjusts the GPT prompt accordingly — no separate workflow needed for different episode formats ✅ FULL_SHOW_NOTES compiled in one step — All eight sections are assembled into a single copy-paste-ready document — saved in its own column so you can paste the entire thing into Spotify or your website without reassembling ✅ Exact timestamps extracted from the transcript — Highlights use the real MM:SS values from the transcript — GPT is instructed not to invent or approximate timestamps ✅ Resources only extracted from what was actually said — The prompt explicitly forbids adding generic or invented resources — you only get links and tools the host or guest mentioned in the recording ✅ Episode duration calculated and formatted — The total episode length in MM:SS is computed from the last transcript segment and logged in the sheet — useful for show notes headers and episode descriptions ✅ Word count logged per episode — The full show notes word count is saved in the sheet so you can verify length before publishing ✅ Guest bio based only on transcript content — If it is an interview episode, GPT writes the bio from what was actually said in the recording — not invented credentials Customisation Options Add a retry limit to stop infinite polling — Before node 6. Wait — 30 Seconds Retry, add a Set step that increments a poll counter, then add a second IF check to stop after 15 polls and send a Gmail error notification to the host instead of looping indefinitely. Send the show notes to Gmail for review — After node 11. Google Sheets — Save Show Notes Library, add a Gmail step that sends the episode title, SEO description, key takeaways, and the full show notes body to a review email address so the host can check before publishing. Add a Slack notification when show notes are ready — After node 11. Google Sheets — Save Show Notes Library, add a Slack step that posts the episode number, title, guest name (or "Solo"), and a link to the Show Notes Library sheet to a #podcast-team channel. Increase the number of timestamped highlights — In the system prompt of node 8. AI Agent — Write Show Notes, change 6-8 key moments to 8-10 key moments for longer episodes with more discussion points worth highlighting. Change the SEO description length — In the system prompt of node 8. AI Agent — Write Show Notes, change 150-200 word episode description to a different range — for example 200-250 words for more detailed descriptions or 100-150 words for shorter platform-style summaries. Troubleshooting Form submission not starting the workflow: Confirm the workflow is Active — inactive workflows do not receive form submissions Copy the Form URL fresh from node 1. Form — Episode Recording + Details after activating Make sure all required fields are filled in — Guest Name and Guest Title / Company are optional; all other fields are required WayinVideo API key errors: Confirm YOUR_WAYINVIDEO_API_KEY in node 2. WayinVideo — Submit Transcription is replaced with your actual key — this workflow uses /v2/transcripts, confirm the URL in node 2 is correct Confirm the same replacement was made in node 4. WayinVideo — Get Transcript Results — both steps require the key Check the execution log of node 2 for the raw error — a 401 means wrong key, a 422 means the URL format is not supported by WayinVideo Workflow stuck in the polling loop: Check that the recording URL is publicly accessible — private Zoom links, expired recordings, or login-required videos will not be transcribed Open the execution log of node 4. WayinVideo — Get Transcript Results and check the raw response status — WayinVideo may have returned FAILED with a specific reason Long episodes (over 60 minutes) may require many polling cycles before completing — this is expected; the loop continues automatically GPT not generating all eight sections or show notes missing: Confirm the API key is connected in node 9. OpenAI — GPT-4o-mini Model and your account has available credits If EPISODE_TITLE, SEO_DESCRIPTION, or FULL_SHOW_NOTES are empty, node 10. Code — Parse Show Notes Output throws a clear error — check the execution log of node 8. AI Agent — Write Show Notes for the raw GPT output to see which section is missing For very long episodes with dense transcripts, the input may approach GPT's token limit — this is rare but can be resolved by shortening the transcript in step 7 Google Sheets not saving the row: Confirm YOUR_GOOGLE_SHEET_ID in node 11. Google Sheets — Save Show Notes Library is replaced with your actual Sheet ID Confirm the tab is named Show Notes Library exactly and all 19 column headers in row 1 match exactly Check that the Google Sheets OAuth2 credential is connected and not expired — re-authorize if needed Support Need help setting this up or want a custom version built for your team or agency? 📧 Email: info@isawow.com 🌐 Website: https://isawow.com

⚠️ Community Node Disclaimer This template uses the Apify LinkedIn Profile Scraper, which is a community node only available in self-hosted n8n installations. The LinkedIn scraping step is optional and can be removed for n8n Cloud compatibility. Who's it for Sales and marketing teams processing 20+ leads daily who need to eliminate manual research and focus reps on hot prospects. Perfect for B2B companies wanting to qualify inbound leads at scale using AI-powered enrichment and scoring. What it does This workflow automates lead qualification by enriching email addresses with firmographic data from People Data Labs, researching individuals and companies using Perplexity AI, scoring leads against your ICP criteria with Claude, and routing them to appropriate channels. Hot leads (8-10 score) get instant Slack alerts with personalized email drafts. Warm leads (5-7) go to a digest channel. Cold leads (0-4) log to your CRM only. Processing takes 30-60 seconds per lead versus 20 minutes manual research, costing $0.08-0.15 per lead. How it works The webhook receives an email address and optional name. Multiple enrichment sources run in parallel: PDL fetches contact and firmographic data, Perplexity researches the individual's recent activity and company developments, and optionally Apify scrapes their LinkedIn profile. All data merges into a complete profile. Claude AI scores the lead against your ICP rules stored in Google Docs, calculating points for company fit, title fit, buying signals, and timing. Based on the total score, leads route to three tiers with different handling. Hot leads trigger immediate Slack alerts and generate personalized email drafts using Gemini. All qualified leads optionally sync to your CRM. Requirements People Data Labs API (or Apollo/Clearbit alternative) Perplexity API Anthropic Claude API Google Docs for ICP rules Slack workspace Gmail account Optional: Apify for LinkedIn scraping (self-hosted only) Optional: HubSpot or other CRM Set up steps 1. Configure the webhook In the Webhook node, set your webhook path (default is "lead-intake"). Send POST requests with this JSON format: { "email": "lead@company.com", "name": "Optional Name" } 2. Add API credentials securely People Data Labs: In the PDL Enrich node, click "Credential for Header Auth" → Create new credential → Add header name X-Api-Key with your PDL API key as the value. This uses n8n's credential management instead of hardcoding keys. Perplexity: In both Individual Research and Company Research nodes, add your Perplexity API credentials. Anthropic: In the Anthropic Chat Model node, add your Claude API credentials. Slack: In both Slack nodes, set up OAuth2 and select your target channels. Hot and warm leads can route to different channels. Gmail: In the Send Hot Lead Email node, configure OAuth2 credentials. Google Docs: In the ICP & Use Case node, replace the documentURL with your Google Doc containing ICP scoring rules, then add OAuth2 credentials. Optional - Apify: In LinkedIn Profile Scraper node, add your Apify OAuth2 credentials from https://apify.com/curious_coder/linkedin-profile-scraper Optional - HubSpot: Enable the Upsert to HubSpot CRM node and add your credentials. Customize the customPropertiesValues array to match your fields. 3. Create your ICP rules document Create a Google Doc with this structure: COMPANY FIT (0-3 points): Company size: 50-500 employees = 3 points Industry: SaaS/Technology = 3 points Geography: North America = 3 points TITLE FIT (0-3 points): VP/C-level = 3 points Director = 2 points Manager = 1 point BUYING SIGNALS (0-2 points): Recent funding = 2 points New executive = 1 point TIMING (0-2 points): Urgent need = 2 points Copy the URL and paste it in the ICP & Use Case node's documentURL parameter. 4. Test the workflow Activate the workflow and send a test webhook. Monitor the execution to verify enrichment sources return data, AI scoring completes, routing works correctly, and notifications send to the right channels. How to customize Swap enrichment sources: Replace the PDL Enrich node with Apollo or Clearbit HTTP Request nodes. Update the Merge Enrichment Data node to parse the new response format. Adjust scoring thresholds: In the AI Agent node prompt, change the score ranges (currently 8-10 = hot, 5-7 = warm, 0-4 = cold) and add custom scoring factors like technology stack match or budget authority. Change routing: In the Route by Score node, add new output conditions for additional tiers like VIP or modify existing thresholds. Different notifications: Replace Slack nodes with Gmail or add Twilio nodes for SMS. Update the formatting nodes to create appropriate message templates. Use different AI models: Swap the Anthropic Chat Model with OpenAI for GPT-4 or replace the Gemini formatting nodes with Claude for consistency. Remove LinkedIn scraping: Delete the LinkedIn Profile Scraper node and adjust Merge All Sources to accept 4 inputs instead of 5 for n8n Cloud compatibility. Connect different CRMs: Replace the HubSpot node with Salesforce, Pipedrive, or other CRM nodes. Update the Format for CRM node's field mappings to match your CRM's structure.

Description Automate your weekly cross-platform social media analytics workflow with AI-powered insights. 📊🤖 This system retrieves real-time Twitter (X) and Facebook data, validates and merges the metrics, formats them via custom JavaScript, generates a visual HTML summary with GPT-4o, stores structured analytics in Notion, and broadcasts key results through Gmail and Slack — all in one seamless flow. Perfect for marketing, social media, and growth teams tracking weekly engagement trends. 🚀💬 What This Template Does 1️⃣ Starts on manual execution to fetch the latest performance data. 🕹️ 2️⃣ Collects live metrics from both Twitter (X API) and Facebook Graph API. 🐦📘 3️⃣ Merges API responses into one unified dataset for analysis. 🧩 4️⃣ Validates data completeness before processing; logs missing or invalid data to Google Sheets. 🔍 5️⃣ Uses JavaScript to normalize data into clean JSON structures for AI analysis. 💻 6️⃣ Leverages Azure OpenAI GPT-4o to generate a professional HTML analytics report. 🧠📈 7️⃣ Updates Notion’s “Growth Chart” database with historical metrics for record-keeping. 🗂️ 8️⃣ Sends the HTML report via Gmail to the marketing or analytics team. 📧 9️⃣ Posts a summarized Slack message highlighting key insights and platform comparisons. 💬 Key Benefits ✅ Eliminates manual social media reporting with full automation. ✅ Ensures clean, validated data before report generation. ✅ Delivers visually engaging HTML performance summaries. ✅ Centralizes analytics storage in Notion for trend tracking. ✅ Keeps teams aligned with instant Slack and Gmail updates. Features Dual-platform analytics integration (Twitter X + Facebook Graph). Custom JavaScript node for data normalization and mapping. GPT-4o model integration for HTML report generation. Real-time error logging to Google Sheets for transparency. Notion database update for structured performance tracking. Slack notifications with emoji-rich summaries and insights. Gmail automation for formatted weekly performance emails. Fully modular — easy to scale to other social platforms. Requirements Twitter OAuth2 API credentials for fetching X metrics. Facebook Graph API credentials for retrieving page data. Azure OpenAI credentials for GPT-4o AI report generation. Notion API credentials with write access to “Growth Chart.” Slack Bot Token with chat:write permission for updates. Google Sheets OAuth2 credentials for error logs. Gmail OAuth2 credentials to send HTML reports. Environment Variables TWITTER_API_KEY FACEBOOK_GRAPH_TOKEN AZURE_OPENAI_KEY NOTION_GROWTH_DB_ID SLACK_ALERT_CHANNEL_ID GOOGLE_SHEET_ERROR_LOG_ID GMAIL_MARKETING_RECIPIENTS Target Audience 📈 Marketing and growth teams analyzing engagement trends. 💡 Social media managers tracking cross-channel performance. 🧠 Data and insights teams needing AI-based summaries. 💬 Brand strategists and content teams monitoring audience health. 🧾 Agencies and operations teams automating weekly reporting. Step-by-Step Setup Instructions 1️⃣ Connect all required API credentials (Twitter, Facebook, Azure OpenAI, Notion, Gmail, Slack, Sheets). 2️⃣ Replace the username and page IDs in the HTTP Request nodes for your brand handles. 3️⃣ Verify the JavaScript node output structure for correct field mapping. 4️⃣ Configure the Azure GPT-4o prompt with your preferred tone and formatting. 5️⃣ Link your Notion database and confirm property names match (followers, likes, username). 6️⃣ Add recipient email(s) in the Gmail node. 7️⃣ Specify your Slack channel ID for automated alerts. 8️⃣ Test run the workflow manually to validate end-to-end execution. 9️⃣ Activate or schedule the workflow for regular weekly reporting. ✅

Quick overview This workflow runs daily to fetch multiple Google Alerts RSS feeds, uses OpenAI GPT-4.1-mini to score and summarize each alert with sentiment and recommended actions, then compiles the most relevant items into a branded HTML digest and sends it via Gmail. How it works Runs every 24 hours on a schedule. Loads a list of Google Alerts RSS feed URLs (with a name and category for each) and fetches each feed as XML. Parses each feed’s Atom entries from the last 48 hours and extracts the title, link, summary, published date, and feed metadata. Skips any feeds that have no recent entries. Sends each parsed alert to OpenAI (GPT-4.1-mini) to generate a relevance score (1–10), one-sentence insight, sentiment, and recommended action. Keeps only alerts with a relevance score of 4 or higher, aggregates them into a single list, and builds a grouped, sorted HTML email digest. Sends the HTML digest email via Gmail with a subject line that reflects the number of relevant alerts. Setup Get your Google Alerts RSS feed URL. Go to google.com/alerts, find an existing alert, click the RSS icon at the bottom, and copy the feed URL. Open node 2. Code — Set Alert Feed URLs and replace YOUR_ALERT_NAME with a descriptive name for the alert, replace YOUR_GOOGLE_ALERTS_RSS_URL with the RSS URL you just copied, and set the category value to any label you want (for example Brand Monitoring or Competitor Intelligence). To add more feeds, duplicate the object inside the array following the commented examples in the code. Open the OpenAI — GPT-4.1-mini Model step and connect your OpenAI API credential. Open node 11. Gmail — Send Digest Email, connect your Gmail OAuth2 credential, and replace REPLACE_WITH_YOUR_EMAIL@example.com with your actual email address. Open node 10. Code — Build HTML Email Digest and find YOUR_BRAND_NAME near the bottom of the code in the footer line — replace it with your company or personal name. Activate the workflow. It will run automatically every 24 hours. To test immediately, use the manual Execute option on node 1. Schedule — Every 24 Hours. Requirements Active n8n instance (self-hosted or cloud) One or more Google Alerts set up at google.com/alerts with RSS feed enabled OpenAI account with GPT-4.1-mini API access Gmail account connected via OAuth2 for sending the digest No additional accounts or API keys required — Google Alerts RSS feeds are publicly accessible Customization Add more Google Alerts feeds — in node 2. Code — Set Alert Feed URLs, duplicate the object inside the alertFeeds array and add any additional RSS feed URL with its own name and category; each feed is fetched and analyzed independently every day Lower or raise the relevance filter threshold — in node 8. IF — Filter Relevance (Score 4 or Above), change the value from 4 to any number between 1 and 10 to include more or fewer alerts in the daily digest Change the digest frequency — in node 1. Schedule — Every 24 Hours, change the interval from 24 hours to 12 hours for twice-daily digests or to 48 hours for every-other-day delivery Add a Slack notification for high-priority alerts — after node 9. Aggregate — Collect All Alerts, add a Slack step that filters for alerts with analysis_relevance of 8 or above and posts a quick message to a brand monitoring channel Extend the article lookback window — in node 4. Code — Parse RSS Entries, change the 48 in the hoursDiff comparison to 72 or 96 to catch alerts from the past 3 or 4 days — useful if you run the workflow less frequently Additional info The Google Alerts RSS feed URL must be copied exactly from your Google Alerts dashboard. Go to google.com/alerts, find your alert, and look for the RSS feed icon at the bottom of the alert row. If you do not see the RSS icon, make sure you are signed into your Google account and the alert is set to deliver via RSS. The workflow parses Atom-format XML which is the format Google Alerts uses for its RSS feeds. Standard RSS XML from other sources uses a different tag structure and may not parse correctly with this workflow without modifying node 4. Code — Parse RSS Entries. The relevance filter at node 8. IF — Filter Relevance (Score 4 or Above) means alerts GPT scores as 1, 2, or 3 are dropped and never appear in the email. If your digest is empty today, it may mean all alerts scored below 4 — you can temporarily lower the threshold to 1 in that node to test that feeds and parsing are working correctly. Replace YOUR_BRAND_NAME in node 10. Code — Build HTML Email Digest — this text appears in the email footer and in the subject line that recipients see. Leaving the placeholder unchanged will show YOUR_BRAND_NAME in every email you receive.

Simplify financial oversight with this automated n8n workflow. Triggered daily, it fetches cash flow and expense data from a Google Sheet, analyzes inflows and outflows, validates records, and generates a comprehensive daily report. The workflow sends multi-channel notifications via email and Slack, ensuring finance professionals stay updated with real-time financial insights. 💸📧 Key Features Daily automation keeps cash flow tracking current. Analyzes inflows and outflows for actionable insights. Multi-channel alerts enhance team visibility. Logs maintain a detailed record in Google Sheets. Workflow Process The Every Day node triggers a daily check at a set time. Get Cash Flow Data** retrieves financial data from a Google Sheet. Analyze Inflows & Outflows** processes the data to identify trends and totals. Validate Records** ensures all entries are complete and accurate. If records are valid, it branches to: Sends Email Daily Report to finance team members. Send Slack Alert to notify the team instantly. Logs to Sheet** appends the summary data to a Google Sheet for tracking. Setup Instructions Import the workflow into n8n and configure Google Sheets OAuth2 for data access. Set the daily trigger time (e.g., 9:00 AM IST) in the "Every Day" node. Test the workflow by adding sample cash flow data and verifying reports. Adjust analysis parameters as needed for specific financial metrics. Prerequisites Google Sheets OAuth2 credentials Gmail API Key for email reports Slack Bot Token (with chat:write permissions) Structured financial data in a Google Sheet Google Sheet Structure: Create a sheet with columns: Date Cash Inflow Cash Outflow Category Notes Updated At Modification Options Customize the "Analyze Inflows & Outflows" node to include custom financial ratios. Adjust the "Validate Records" filter to flag anomalies or missing data. Modify email and Slack templates with branded formatting. Integrate with accounting tools (e.g., Xero) for live data feeds. Set different trigger times to align with your financial review schedule. Discover more workflows – Get in touch with us

🚀 Overview This n8n workflow automatically generates professional API documentation from C header (.h) files using AI. It scans a Google Drive folder for header files, extracts the source code, sends it to GPT-4o for structured analysis, and generates a beautiful HTML documentation page. The final documentation is uploaded back to Google Drive and a completion email is sent. This workflow is ideal for embedded systems teams, firmware engineers, and SDK developers who want an automated documentation pipeline. ✨ Key Features ⚡ Fully automated documentation generation 📁 Reads .h files directly from Google Drive 🤖 Uses AI to analyze C APIs and extract documentation 📑 Generates clean HTML documentation 📊 Documents functions, types, enums, and constants 🔁 Processes files one-by-one for reliability ☁️ Saves generated documentation back to Google Drive 📧 Sends a completion email notification 🧠 What the AI Extracts The workflow automatically identifies and documents: 📘 Overview of the header file 🔧 Functions Signatures Parameters Return values Usage examples 🧩 Enumerations 🧱 Data Types & Structures 🔢 Constants / Macros 📝 Developer Notes 🖥 Generated Documentation The output is a clean developer-friendly HTML documentation page including: 🧭 Sidebar navigation 📌 Function cards 📊 Parameter tables 💻 Code examples 🎨 Professional developer layout Perfect for: Developer portals SDK documentation Internal engineering documentation Embedded system libraries ⚙️ Workflow Architecture | Step | Node | Purpose | |-----|-----|--------| | 1 | ▶️ Manual Trigger | Starts the workflow | | 2 | 📂 Get all files | Reads files from Google Drive | | 3 | 🔎 Filter .h files | Keeps only header files | | 4 | 🔁 Split in Batches | Processes files sequentially | | 5 | ⬇️ Download file | Downloads the header file | | 6 | 📖 Extract text | Extracts code content | | 7 | 🤖 AI Extraction | AI extracts API structure | | 8 | 🧹 Parse JSON | Cleans AI output | | 9 | 🎨 Generate HTML | Builds documentation page | |10 | ☁️ Upload to Drive | Saves documentation | |11 | 📧 Email notification | Sends completion email | 🔧 Requirements To run this workflow you need: 🔹 Google Drive OAuth2 credentials 🔹 OpenAI API credentials 🔹 Gmail credentials 🛠 Setup Guide 1️⃣ Configure Google Drive Create two folders. Source folder Output folder Update the folder IDs in the nodes: Get all files from folder Save documentation to Google Drive 2️⃣ Configure OpenAI Add an OpenAI credential in n8n. Model used: The model analyzes C header files and returns structured API documentation. 3️⃣ Configure Gmail Add a Gmail OAuth credential. Update the recipient address inside: ▶️ Run the Workflow Click Execute Workflow. The workflow will: 1️⃣ Scan the Google Drive folder 2️⃣ Process each .h file 3️⃣ Generate HTML documentation 4️⃣ Upload documentation to Drive 5️⃣ Send a completion email 🖼 Documentation Preview 💡 Use Cases 🔧 Embedded firmware documentation 📦 SDK documentation generation 🧑‍💻 Developer portal automation 📚 C library documentation ⚙️ Continuous documentation pipelines 🔮 Future Improvements This workflow can be extended with several enhancements: 📄 PDF Documentation Export Add a step to convert the generated HTML documentation into PDF files using tools such as: Puppeteer HTML-to-PDF services n8n community PDF nodes This allows teams to distribute documentation as downloadable reports. 🔐 Local AI for Security (Ollama / Open-Source Models) Instead of using the OpenAI node, the workflow can be modified to run fully locally using AI models such as: Ollama** Open-source LLMs (Llama, Mistral, CodeLlama)** These models can run on your own server, which provides: 🔒 Better data privacy 🏢 No external API calls ⚡ Faster responses on local infrastructure 🛡 Increased security for proprietary source code This can be implemented in n8n using: HTTP Request node → Ollama API** Local AI inference servers Private LLM deployments 📚 Multi-Language Documentation The workflow could also support additional languages such as: .c .cpp .hpp .rs .go

This is a production-ready, end-to-end workflow that automatically compares hotel prices across multiple booking platforms and delivers beautiful email reports to users. Unlike basic building blocks, this workflow is a complete solution ready to deploy. ✨ What Makes This Production-Ready ✅ Complete End-to-End Automation Input**: Natural language queries via webhook Processing**: Multi-platform scraping & comparison Output**: Professional email reports + analytics Feedback**: Real-time webhook responses ✅ Advanced Features 🧠 Natural Language Processing for flexible queries 🔄 Parallel scraping from multiple platforms 📊 Analytics tracking with Google Sheets integration 💌 Beautiful HTML email reports 🛡️ Error handling and graceful degradation 📱 Webhook responses for real-time feedback ✅ Business Value For Travel Agencies**: Instant price comparison service for clients For Hotels**: Competitive pricing intelligence For Travelers**: Save time and money with automated research 🚀 Setup Instructions Step 1: Import Workflow Copy the workflow JSON from the artifact In n8n, go to Workflows → Import from File/URL Paste the JSON and click Import Step 2: Configure Credentials A. SMTP Email (Required) Settings → Credentials → Add Credential → SMTP Host: smtp.gmail.com (for Gmail) Port: 587 User: your-email@gmail.com Password: your-app-password (not regular password!) Gmail Setup: Enable 2FA on your Google Account Generate App Password: https://myaccount.google.com/apppasswords Use the generated password in n8n B. Google Sheets (Optional - for analytics) Settings → Credentials → Add Credential → Google Sheets OAuth2 Follow the OAuth flow to connect your Google account Sheet Setup: Create a new Google Sheet Name the first sheet "Analytics" Add headers: timestamp, query, hotel, city, checkIn, checkOut, bestPrice, platform, totalResults, userEmail Copy the Sheet ID from URL and paste in the "Save to Google Sheets" node Step 3: Set Up Scraping Service You need to create a scraping API that the workflow calls. Here are your options: Option A: Use Your Existing Python Script Create a simple Flask API wrapper: api_wrapper.py from flask import Flask, request, jsonify import subprocess import json app = Flask(name) @app.route('/scrape/<platform>', methods=['POST']) def scrape(platform): data = request.json query = f"{data['checkIn']} to {data['checkOut']}, {data['hotel']}, {data['city']}" try: result = subprocess.run( ['python3', 'price_scrap_2.py', query, platform], capture_output=True, text=True, timeout=30 ) Parse your script output output = result.stdout Assuming your script returns price data return jsonify({ 'price': extracted_price, 'currency': 'USD', 'roomType': 'Standard Room', 'url': booking_url, 'availability': True }) except Exception as e: return jsonify({'error': str(e)}), 500 if name == 'main': app.run(host='0.0.0.0', port=5000) Deploy: pip install flask python api_wrapper.py Update n8n HTTP Request nodes: URL: http://your-server-ip:5000/scrape/booking URL: http://your-server-ip:5000/scrape/agoda URL: http://your-server-ip:5000/scrape/expedia Option B: Use Third-Party Scraping Services Recommended Services: ScraperAPI** (scraperapi.com) - $49/month for 100k requests Bright Data** (brightdata.com) - Pay as you go Apify** (apify.com) - Has pre-built hotel scrapers Example with ScraperAPI: // In HTTP Request node URL: http://api.scraperapi.com Query Parameters: api_key: YOUR_API_KEY url: https://booking.com/search?hotel={{$json.hotelName}}... Option C: Use n8n SSH Node (Like Your Original) Keep your SSH approach but improve it: Replace HTTP Request nodes with SSH nodes Point to your server with the Python script Ensure error handling and timeouts // SSH Node Configuration Host: your-server-ip Command: python3 /path/to/price_scrap_2.py "{{$json.hotelName}}" "{{$json.city}}" "{{$json.checkInISO}}" "{{$json.checkOutISO}}" "booking" Step 4: Activate Webhook Click on "Webhook - Receive Request" node Click "Listen for Test Event" Copy the webhook URL (e.g., https://your-n8n.com/webhook/hotel-price-check) Test with this curl command: curl -X POST https://your-n8n.com/webhook/hotel-price-check \ -H "Content-Type: application/json" \ -d '{ "message": "I want to check Marriott Hotel in Singapore from 15th March to 18th March", "email": "user@example.com", "name": "John Doe" }' Step 5: Activate Workflow Toggle the workflow to Active The webhook is now live and ready to receive requests 📝 Usage Examples Example 1: Basic Query { "message": "Hilton Hotel in Dubai from 20th December to 23rd December", "email": "traveler@email.com", "name": "Sarah" } Example 2: Flexible Format { "message": "I need prices for Taj Hotel, Mumbai. Check-in: 5th January, Check-out: 8th January", "email": "customer@email.com" } Example 3: Short Format { "message": "Hyatt Singapore March 10 to March 13", "email": "user@email.com" } 🎨 Customization Options 1. Add More Booking Platforms Steps: Duplicate an existing "Scrape" node Update the platform parameter Connect it to "Aggregate & Compare" Update the aggregation logic to include the new platform 2. Change Email Template Edit the "Format Email Report" node's JavaScript: Modify HTML structure Change colors (currently purple gradient) Add your company logo Include terms and conditions 3. Add SMS Notifications Using Twilio: Add new node: Twilio → Send SMS Connect after "Aggregate & Compare" Format: "Best deal: ${hotel} at ${platform} for ${price}" 4. Add Slack Integration Add Slack node after "Aggregate & Compare" Send to #travel-deals channel Include quick booking links 5. Implement Caching Add Redis or n8n's built-in cache: // Before scraping, check cache const cacheKey = ${hotelName}-${city}-${checkIn}-${checkOut}; const cached = await $cache.get(cacheKey); if (cached && Date.now() - cached.timestamp < 3600000) { return cached.data; // Use 1-hour cache } 📊 Analytics & Monitoring Google Sheets Dashboard The workflow automatically logs to Google Sheets. Create a dashboard with: Metrics to track: Total searches per day/week Most searched hotels Most searched cities Average price ranges Platform with best prices (frequency) User engagement (repeat users) Example Sheet Formulas: // Total searches today =COUNTIF(A:A, TODAY()) // Most popular hotel =INDEX(C:C, MODE(MATCH(C:C, C:C, 0))) // Average best price =AVERAGE(G:G) Set Up Alerts Add a node after "Aggregate & Compare": // Alert if prices are unusually high if (bestDeal.price > avgPrice * 1.5) { // Send alert to admin return [{ json: { alert: true, message: High prices detected for ${hotelName} } }]; } 🛡️ Error Handling The workflow includes comprehensive error handling: 1. Missing Information If user doesn't provide hotel/city/dates → Responds with helpful prompt 2. Scraping Failures If all platforms fail → Sends "No results" email with suggestions 3. Partial Results If some platforms work → Shows available results + notes errors 4. Email Delivery Issues Uses continueOnFail: true to prevent workflow crashes 🔒 Security Best Practices 1. Rate Limiting Add rate limiting to prevent abuse: // In Parse & Validate node const userEmail = $json.email; const recentSearches = await $cache.get(searches:${userEmail}); if (recentSearches && recentSearches.length > 10) { return [{ json: { status: 'rate_limited', response: 'Too many requests. Please try again in 1 hour.' } }]; } 2. Input Validation Already implemented - validates hotel names, cities, dates 3. Email Verification Add email verification before first use: // Send verification code const code = Math.random().toString(36).substring(7); await $sendEmail({ to: userEmail, subject: 'Verify your email', body: Your code: ${code} }); 4. API Key Protection Never expose scraping API keys in responses or logs 🚀 Deployment Options Option 1: n8n Cloud (Easiest) Sign up at n8n.cloud Import workflow Configure credentials Activate Pros: No maintenance, automatic updates Cons: Monthly cost Option 2: Self-Hosted (Most Control) Using Docker docker run -it --rm \ --name n8n \ -p 5678:5678 \ -v ~/.n8n:/home/node/.n8n \ n8nio/n8n Using npm npm install -g n8n n8n start Pros: Free, full control Cons: You manage updates Option 3: Cloud Platforms Railway.app (recommended for beginners) DigitalOcean App Platform AWS ECS Google Cloud Run 📈 Scaling Recommendations For < 100 searches/day Current setup is perfect Use n8n Cloud Starter or small VPS For 100-1000 searches/day Add Redis caching (1-hour cache) Use queue system for scraping Upgrade to n8n Cloud Pro For 1000+ searches/day Implement job queue (Bull/Redis) Use dedicated scraping service Load balance multiple n8n instances Consider microservices architecture 🐛 Troubleshooting Issue: Webhook not responding Solution: Check workflow is Active Verify webhook URL is correct Check n8n logs: Settings → Log Streaming Issue: No prices returned Solution: Test scraping endpoints individually Check if hotel name matches exactly Verify dates are in future Try different date ranges Issue: Emails not sending Solution: Verify SMTP credentials Check "less secure apps" setting (Gmail) Use App Password instead of regular password Check spam folder Issue: Slow response times Solution: Enable parallel scraping (already configured) Add timeout limits (30 seconds recommended) Implement caching Use faster scraping service

This enterprise-grade n8n workflow automates the entire event planning lifecycle — from client briefs to final reports — using Claude AI, real-time financial data, and smart integrations. It converts raw client data into optimized, insight-driven event plans with cost savings, risk management, and automatic reporting, all with zero manual work. Key Features Multi-source data fusion** from Google Sheets (ClientBriefs, BudgetEstimates, ActualCosts, VendorDatabase) AI-powered orchestration* using *Claude 3.5 Sonnet** for event plan optimization Automatic ROI and variance analysis** with cost-saving insights Vendor intelligence** — ranks suppliers by cost, rating, and reliability Risk engine** computes event risk (probability × impact) Auto-approval logic** for safe, high-ROI events Multi-channel delivery:** Slack + Email + Google Sheets Audit-ready:** Full JSON plan + execution logs Scalable triggers:** Webhook or daily schedule Workflow Process | Step | Node | Description | | ---- | --------------------------- | -------------------------------------------------------- | | 1 | Orchestrate Trigger | Runs daily at 7 AM or via webhook (/event-orchestrate) | | 2 | Read Client Brief | Loads event metadata from the ClientBriefs sheet | | 3 | Read Budget Estimates | Fetches estimated budgets and vendor data | | 4 | Read Actual Costs | Loads live cost data for comparison | | 5 | Read Vendor Database | Pulls vendor pricing, reliability, and rating | | 6 | Fuse All Data | Merges data into a unified dataset | | 7 | Data Fusion Engine | Calculates totals, variances, and validates inputs | | 8 | AI Orchestration Engine | Sends structured prompt to Claude AI for analysis | | 9 | Parse & Finalize | Extracts JSON, computes ROI, risks, and savings | | 10 | Save Orchestrated Plan | Updates OrchestratedPlans sheet with results | | 11 | Team Sync | Sends status & summary to Slack | | 12 | Executive Report | Emails final interactive plan to event planner | Setup Instructions 1. Import Workflow Open n8n → Workflows → Import from Clipboard Paste the JSON workflow 2. Configure Credentials | Integration | Details | | ----------------- | -------------------------------------------------- | | Google Sheets | Service account with spreadsheet access | | Claude AI | Anthropic API key for claude-3-5-sonnet-20241022 | | Slack | Webhook or OAuth app | | Email | SMTP or Gmail OAuth credentials | 3. Update Spreadsheet IDs Ensure your Google Sheets include: ClientBriefs BudgetEstimates ActualCosts VendorDatabase OrchestratedPlans 4. Set Triggers Webhook:** /webhook/event-orchestrate Schedule:** Daily at 7:00 AM 5. Run a Test Use manual execution to confirm: Sheet updates Slack notifications Email delivery Google Sheets Structure ClientBriefs | eventId | clientName | eventType | attendees | budget | eventDate | plannerEmail | spreadsheetId | teamChannel | priority | |----------|-------------|------------|-----------|----------|------------|---------------|---------------|-------------| | EVT-2025-001 | Acme Corp | Conference | 200 | 75000 | 2025-06-15 | sarah@acme.com | 1A... | #event-orchestration | High | BudgetEstimates | category | item | budgetAmount | estimatedCost | vendor | | -------- | -------------- | ------------ | ------------- | ----------- | | Venue | Grand Ballroom | 20000 | 22500 | Luxe Events | ActualCosts | category | actualCost | | -------- | ---------- | | Venue | 23000 | VendorDatabase | vendorName | category | avgCost | rating | reliability | | ----------- | -------- | ------- | ------ | ----------- | | Luxe Events | Venue | 21000 | 4.8 | High | OrchestratedPlans Automatically filled with: eventId, savings, roi, riskLevel, status, summary, fullPlan (JSON) System Requirements | Requirement | Version/Access | | --------------------- | ---------------------------------------------- | | n8n | v1.50+ (LangChain supported) | | Claude AI API | claude-3-5-sonnet-20241022 | | Google Sheets API | https://www.googleapis.com/auth/spreadsheets | | Slack Webhook | Required for notifications | | Email Service | SMTP, Gmail, or SendGrid | Optional Enhancements Add PDF export for management reports Connect Google Calendar for event scheduling Integrate CRM (HubSpot / Salesforce) for client updates Add interactive Slack buttons for approvals Export results to Notion or Airtable Enable multi-event batch orchestration Add forecasting from past data trends Result: A single automated system that plans, analyzes, and reports events — with full AI intelligence and zero manual work. Explore More AI Workflows: https://www.oneclickitsolution.com/contact-us/

How It Works This workflow automates cost analysis and budget optimization for enterprises managing complex operational expenses. Designed for CFOs, finance teams, and operations managers, it addresses the challenge of identifying cost inefficiencies and generating actionable recommendations in real-time. The system runs every 15 minutes, monitoring cost metrics and generating AI performance data. The Cost Intelligence Agent aggregates financial data before routing to parallel AI processing. Claude AI executes budget optimization analysis while a specialized cost model identifies spending patterns. Routing engines evaluate optimization strategies, with NVIDIA parsers ensuring standardized outputs. The Optimization Coordinator consolidates insights and routes by severity: critical overruns trigger executive alerts via email and Slack, warnings generate management notifications, while routine optimizations proceed to documentation and historical storage for trend analysis. Setup Steps Configure Schedule Trigger for 15-minute intervals Add Claude API credentials in Workflow Configuration and Budget Alert Tool nodes Set up NVIDIA API keys in Cost Intelligence Agent and Structured Output Parser nodes Connect Gmail authentication and configure finance team distribution lists Integrate Slack workspace credentials and specify budget alert channels Configure storage endpoints in cost history nodes with database connections Prerequisites Claude API access, NVIDIA API credentials, Gmail/Google Workspace account, Slack workspace integration Use Cases Multi-department budget variance analysis, cloud cost optimization, procurement pattern detection Customization Integrate ERP systems, add department-specific rules, customize alert thresholds by category Benefits Reduces overruns 40% through early detection, identifies 15-20% monthly savings