I created an n8n workflow to send daily security news to your Discord

Video Tutorial: https://youtu.be/ZSIyHFEgF2I

Workflow File: https://github.com/th3-ch0s3n-0n3/n8n-workflows/blob/main/Daily_Sec_News.json

Overview and Full Details: https://github.com/th3-ch0s3n-0n3/n8n-workflows/blob/main/Readme.md

Overview

This workflow runs on demand (via chat trigger) and performs the following:

1. Posts a "Here are your security news for today" header message to Discord
2. Fetches the latest posts from 4 security-focused RSS feeds (capped at 5 items each)
3. Merges and aggregates the feed data into a single batch
4. Sends the batch to Gemini 2.5 Flash Lite with a detailed Senior CTI Analyst prompt
5. Posts the AI-filtered, formatted findings back to Discord

AI Filtering Logic

The aggregated batch is sent to Gemini 2.5 Flash Lite with a role-prompted CTI analyst system prompt. The model is instructed to use only the provided news batch (no external knowledge) and filter strictly by the following criteria.

Focus Areas (report only these)

• Supply chain attacks: malicious or typosquatted packages on npm, PyPI, RubyGems, crates.io, Maven, NuGet
• Malicious browser extensions targeting developers
• IDE or coding agent attacks (VS Code, JetBrains, Cursor, Copilot, Claude, etc.)
• Critical CVEs (CVSS ≥ 9.0) in widely-used developer tools or libraries
• Credential theft or backdoors via CI/CD tools, GitHub Actions, or build pipelines

Strict Ignore List

• Opinion pieces, editorials, award announcements
• Government breaches or geopolitical events
• Product launches, feature updates, or pricing changes
• Generic phishing/social engineering without a technical package/tool vector
• Company data breaches (unless caused by a supply chain or qualifying CVE)
• CVEs in server-side templating or serialization libraries without confirmed active exploitation
• CVEs where the affected library is not a direct developer dependency

Customization

• Add more feeds — duplicate any RSS node, add a matching Limit node, and wire both into the Merge node (increase numberInputs accordingly).
• Change the item limit — adjust the maxItems value in any Limit node (currently 5 per feed).
• Swap the model — replace the Gemini node with any other LangChain-compatible model node in n8n.
• Change the output channel — update guildId and channelId in the Discord nodes.
• Adjust focus areas — edit the system prompt in the "Message a model" node to narrow or expand CTI scope.