Multi-Source News Curator with Mistral AI Analysis, Summaries & Custom Channels

Download [51.4KB]

Nodes

+25

Flexible News Curator - Multi-Sources, AI Analysis, Summaries, Translation, and Settable Channels

🎬 Overview

The Flexible News Curator workflow can automate the collection, filtering, AI-driven analysis, and summarization of news from diverse sources of your interest. Using customizable search themes, RSS feeds, and (optional) video descriptions, it delivers concise, quality news summaries via configurable channels. This workflow is designed to help reduce information overload and keep you updated effortlessly.

Click the image below to watch the video guide:

✨ Features

Multi-Source News Aggregation: Collect news from customizable RSS feeds, SerpAPI, and Video Channel Feeds (if enabled).
AI-Powered News Selection & Summarization: Uses advanced AI agents (Mistral Cloud Chat Model by default) to select, analyze, and summarize top news.
Quality Assurance Step: Optional AI-powered filtering to improve news selection quality before analysis.
Multi-Language Translation & Tone Customization: Translate summaries and customize tone for localized or tailored consumption.
Multi-Channel Delivery: Send outputs via Email, Telegram, WhatsApp, Webhook, or save to disk.
Advanced Filtering: Regex-based filtering on URLs, titles, and content to exclude unwanted articles.
Sub-Workflow Architecture: Modular handling of video transcripts, content retrieval, multi-theme searching, and more.
Flexible Scheduling & Trigger Options: Supports schedule-based triggering, email (IMAP) triggers, and webhook-based activation.
(Optional) Video Search: Video content descriptions via Video Channel Feeds.

👤 Who is this for?

This workflow benefits professionals, researchers, marketers, and anyone who needs to stay informed about specific news themes without wasting time on irrelevant information or reading too many news to select the most interesting ones.

💡 What problem does this solve?

The workflow tackles the challenge of information overload by automatically filtering, summarizing, and delivering the essential news tailored to your interests and preferences. It integrates various data sources and channels for comprehensive yet efficient news consumption. Ideal use-cases include:

Monitoring breakthroughs in research fields
Receiving daily business opportunity updates (e.g., real estate)
Lower the cognitive load required to follow your favorite news
Translating news summaries to your chosen language

🔍 What this workflow does

The workflow gathers news from RSS feeds, search engines, and social media, then:

Filters duplicates and irrelevant content via custom regex filters and date ranges.
Applies optional AI-powered Quality Assurance for headline evaluation.
Selects top news articles with AI analysis focused on user-defined criteria and audience.
Summarizes individual articles using AI summarization agents, ensuring structured, consistent outputs.
Optionally translates and adjusts the tone of summaries.
Distributes summaries through configured channels such as email, social media, messaging apps, or webhook calls.

🔄 Workflow Steps

1. News Gathering

Fetch news using RSS feeds, SerpAPI search, and optionally video channel feeds.
Standardize output structures for seamless merging.
Employ sub-workflows for video transcript retrieval and looping over custom RSS feed lists.

2. Filtering

Remove duplicates and news outside the specified date range.
Exclude articles matching user-defined keywords via regex filters in URLs, titles, and content.
Limit the number of news articles for AI analysis.

3. News Selection

Optionally invoke an AI Quality Assurance agent to pre-filter headlines.
Aggregate news for AI analysis.
Select and summarize top news articles with AI agents using customizable criteria.
Parse AI responses into defined JSON structures to ensure consistent data.

4. News Summarization

Prepare individual article content.
Summarize content with AI agents and validate structured output.

5. Sender Preparation

Combine general summaries with selected top news summaries.
Format final summaries as text and HTML suitable for delivery.
Optionally apply translation and tone adjustment.

6. Sending

Deliver summaries through selected channels (Email, Telegram, WhatsApp, Webhook).
Optionally save the output to disk as JSON.

📌 Expected Input / Configuration

The workflow is primarily configured via the Configure Workflow Args node or the Global Variables custom node, with these key parameters:

Parameter

Description

Type

search_themes

List of keywords/themes to search in SerpAPI

List of strings

datetime_delta

Number of days back to include news from; e.g., 0 = today

Integer

link_censor

Regex to exclude unwanted URLs

Regex string

title_censor

Regex to exclude unwanted titles

Regex string

content_censor

Regex to exclude unwanted content

Regex string

use_qa

Flag to enable AI Quality Assurance for headline filtering

Boolean

max_news_analysis

Max number of articles sent to News Analyzer

Integer

qa_max_news

Number of headlines the QA Agent analyzes

Integer

qa_max_top_news

Number of headlines selected by QA Agent

Integer

qa_check_criteria

Criteria used by QA Agent to discard low-quality headlines

List of strings

qa_select_criteria

Criteria used by QA Agent to rank/select the best headlines

List of strings

news_focus

What the News Analyzer should focus on while selecting news

String

news_target_audience

Target audience description for the News Analyzer

String

news_criteria

Instructions for the News Analyzer to identify relevant news

List of strings

language

Language for news summaries; triggers translation if not English

String

translator_tone

Tone for translation (e.g., casual, professional)

String

translator_notes

Additional instructions for the translator

String

email_sender

Email address used for sending (via SMTP)

String

email_recipients

Recipient email addresses (comma-separated)

String

email_subject

Email subject line

String

telegram_chat_id

Telegram chat ID for sending notifications

String

phone_number

Phone number for WhatsApp messages

String

rss_feeds

Custom list of RSS feeds (objects with link and needs_content_search properties)

JSON array of objects

video_rss_feeds

Custom list of video RSS feeds

JSON array of objects

enable_video_search

Enable/disable video search functionality

Boolean

enabled_senders

List of enabled delivery channels (email, telegram, whatsapp, webhook, save-to-disk)

List of strings

Hint: To add or combine keywords in censors, use the pattern "keyword1|keyword2|keyword3".

📦 Expected Output

Structured JSON containing a general news summary, top news with summaries, and metadata, suitable for your preferred channel delivery.

📌 Example

An example that includes workflow parameters is provided in a note within the workflow.

⚙️ n8n Setup Used

n8n version: 1.100.1
n8n-nodes-serpapi: 0.1.6
n8n-nodes-globals: 1.1.0
LLM Model: mistral-small-latest (API)
IMAP: imap.gmail.com (Port 993)
Platform: Podman 4.3.1 on Linux
Date: 2025-07-15

⚡ Requirements to Use / Setup

Self-hosted n8n instance.
(This workflow contains community nodes that are only compatible with the self-hosted version of n8n.)
Install necessary custom nodes:
- n8n-nodes-serpapi
- n8n-nodes-globals (or use Edit Field (Set) node instead)
Configure all sub-workflows bundled within this template (see Sub-Workflows Guide).
Provide valid credentials to nodes for SerpAPI, Telegram, WhatsApp, Mistral Cloud Chat API, and SMTP (for email).
Custom RSS feed list must be set by you in the workflow args.
You must either install the SerpAPI custom node or deactivate it.

⚠️ Notes, Assumptions & Warnings

The workflow timeout is set to 30 minutes by default; adjust depending on your setup and workload.
Duplicate removal is applied, but occasional overlaps might still appear depending on feed sources.
This workflow assumes familiarity with n8n, RSS feeds, API key management and regex expressions.
Video search only works for configured video channels; remember to respect the rights of these channels.
Using AI agents (Mistral or substitute LLMs) requires access to their API services and keys.
Out-of-the-box customization is done via the Global Variables node or direct workflow argument edits.

ℹ️ About Us

This workflow was developed by the Hybroht team of AI enthusiasts and developers dedicated to enhancing the capabilities of AI through collaborative processes. Our goal is to create tools that harness the possibilities of AI technology and more. For questions, support, or feature requests, reach out via [email protected].