A daily workflow that filters industry news by relevance using AI and delivers it as a categorized briefing in the team channel, instead of someone scrolling through feeds by hand.
automatingresearchingwriting
Description
The Press Chain delivers a curated industry briefing every morning. The workflow collects the news from the last 24 hours from the stored RSS feeds, uses AI to score each article for relevance to your company, summarizes the relevant hits, and posts the result, sorted by category, in the team channel. Instead of scrolling through feeds yourself, you start the day informed.
The workflow automates source retrieval, relevance scoring against a defined threshold, per-article summarization, and formatting as a structured briefing. The human stays in the loop at two points: defining the company context, meaning industry, core business, competitors and focus topics, and calibrating the relevance threshold based on team feedback after the first runs.
This workflow is useful as an ongoing baseline supply of market and competitor news. The topic clusters that emerge can also serve as input for newsletters, blog posts and press work. It does not replace in-depth competitive research, but keeps the team informed daily without anyone having to search actively.
Daily overview without research: A curated briefing is ready in the team channel at 07:00 each morning.
Faster reaction to market moves: Competitor activity and trends become visible on the same day.
Input for content and press work: The resulting topic clusters can be used directly for newsletters, blog posts and press work.
Less information overload: Only articles above the relevance threshold reach the channel.
Steps
Every step shows who carries it out: icon, colour and label together indicate whether a person acts, whether it runs automatically, whether a result is produced, or whether an approval is required.
01 · Person
The workflow starts daily at a fixed time, for example 7:00 AM, and takes into account news from the last 24 hours.
02 · Automated
All stored RSS feeds are loaded. Title, source, URL and snippet of each article are collected.
03 · Automated
Each article receives an AI relevance score from 1 to 10 based on industry, core business, competitors and focus topics. Only articles at or above a defined threshold, for example 6, move forward.
04 · Automated
For each relevant article, a summary of two to three sentences is created, assigned to a category such as competition, market trends, regulation, technology or funding.
05 · Automated
The categorized summaries are turned into a structured Markdown briefing with a date and clickable links.
06 · Result
The finished briefing is posted in the team channel, with direct links to the original articles.
Key
PersonAutomatedResultApproval
Requirements and operations
Requirements
RSS feed toolFeedly, Inoreader or a comparable tool collects industry news and press releases.
Team chatSlack, Teams or Discord for the news channel.
Context definitionIndustry, core business, competitors and strategic focus topics are clearly defined in advance.
LLM accessAccess to a language model, for example from OpenAI, Anthropic or Google.
Testing
Test runStart the workflow manually and observe the fetch, filter and posting steps.
RSS and filterCheck whether all sources deliver current articles and the threshold works as intended.
Categorization and summaryCheck whether the category is correct and the summary stays concise at two to three sentences.
Edge casesTest a day with no articles, paywalled snippets, duplicates and a large article volume.
Go-live
ApproveActivate the workflow only once sources, filter and posting run reliably.
Calibrate the thresholdStart with five to ten sources and a threshold of at least seven.
Team feedbackFine tune sources and threshold after one to two weeks based on team feedback.
ExpandAdd sentiment analysis, weekly trends or alerts for critical news if needed.
Common issues
No articles in the outputValidate the feed URLs or lower the relevance threshold.
Too many irrelevant articlesSharpen the focus topics in the prompt and raise the threshold.
Posting failsCheck the bot permissions in the workspace and the target channel.
Duplicate articlesAdd a deduplication check on title or URL before the articles reach the language model.
Builder Prompt
Copy the prompt below in full into your automation tool or agent builder. As a file: presse-kette.en.json
GOAL
Build a workflow that collects industry news and press releases from RSS feeds every day, filters them by relevance using AI, summarizes and categorizes the relevant articles, and posts the result as a structured briefing to a team channel.
Adapt the structure to the concepts of your automation tool, for example nodes, steps, zaps or agent actions. While doing so, preserve the trigger logic, the sequence and the data flow.
TRIGGER
Type: time based, daily at a fixed time, for example 7:00 AM.
Scope: articles from the last 24 hours from the configured RSS and news sources, for example Feedly or Inoreader.
STEPS
1. Fetch sources: load all RSS feeds; for each article, collect title, source, URL and snippet.
2. Remove duplicates: remove duplicate articles by title or URL before they reach the language model.
3. AI step, relevance filter: use the relevance filter prompt below exactly. The output is a JSON array; only articles with a relevance_score of 6 or higher move forward.
4. AI step, summary and categorization: use the second prompt below exactly, with the filtered articles as input. The output is the finished Markdown briefing.
5. Output: post the briefing to the Slack or Teams channel, with clickable links to the original articles.
PROMPTS (use exactly as written)
Prompt 1, relevance filter:
"""
You are an expert in market monitoring and analyze press releases for the company every day.
Company context:
Industry: [INDUSTRY]
Core business: [CORE_BUSINESS]
Main competitors: [COMPETITORS]
Strategic focus topics: [FOCUS_TOPICS]
Task:
Analyze the following press releases and news articles and rate their relevance to the company on a scale from 1 to 10.
News articles:
[RSS_FEED_ITEMS]
Relevance criteria:
Score 8 to 10: direct impact on the business, for example competitor activity, new regulations, disruptive technologies.
Score 5 to 7: indirect relevance, for example market trends, industry shifts, customer needs.
Score 1 to 4: low relevance, for example general news, peripheral topics.
Return only articles with a score of 6 or higher.
Output format (JSON):
[
{
"title": "Article title",
"source": "Source",
"url": "Link",
"relevance_score": 8,
"reason": "Brief justification of the relevance"
}
]
"""
Prompt 2, summary and categorization:
"""
You are a business analyst and create a news briefing for the management team every day.
Task:
For each of the following relevant articles, write a concise summary of no more than two to three sentences and assign it to a suitable category.
Relevant articles:
[FILTERED_ARTICLES]
Categories:
Competition: activities of competitors, new market entrants, mergers and acquisitions.
Market trends: industry developments, customer needs, studies.
Regulation: new laws, compliance requirements, policy.
Technology: new tools, technological breakthroughs, innovation.
Funding: funding rounds, IPOs, investor activity.
Output format (Markdown for the team channel):
News briefing, [DATE]
Competition
[Article title] ([Source]([URL]))
Summary in two to three sentences.
Market trends
[Article title] ([Source]([URL]))
Summary in two to three sentences.
[Additional categories]
Automatically generated at [TIMESTAMP]
"""
DATA FLOW
Step 1 to step 3: the collected article list to [RSS_FEED_ITEMS].
Step 3 to step 4: the JSON array of relevant articles to [FILTERED_ARTICLES].
Before step 4: the current date and time for [DATE] and [TIMESTAMP].
Step 4 to step 5: the finished Markdown briefing.
ERROR HANDLING
No article above the threshold: post a short note, for example "No relevant news today", instead of an empty briefing.
A single feed returns nothing or an error: log it and process the remaining sources anyway, then validate the feed URL.
Very large article volumes: pass the articles to the relevance filter in batches.
Paywalled snippets: assess and summarize based on the visible snippet only, without inventing anything.
Posting fails: check bot permissions in the workspace and channel, log the error.
PLACEHOLDER CONVENTION
All placeholders are in square brackets. Mapped dynamically are [RSS_FEED_ITEMS], [FILTERED_ARTICLES], [DATE] and [TIMESTAMP], as described in the data flow. Not dynamic are [INDUSTRY], [CORE_BUSINESS], [COMPETITORS] and [FOCUS_TOPICS]: set these manually before activation. The square brackets in the Markdown output format of prompt 2 are format instructions and stay unchanged. List all placeholders that require manual configuration separately at the end.
SELF CHECK BEFORE COMPLETION
Check, before you report the workflow as finished:
1. Are all steps connected correctly, including the duplicate check before the AI steps?
2. Are both prompts inserted in full and unchanged, and are all dynamic placeholders mapped?
3. Does the relevance threshold of at least 6 points apply, and is the JSON from step 3 parsed correctly?
4. Are all points listed that require manual configuration: feed sources, target channel, bot permissions, the LLM access and the placeholders in the company context?
Working prompts
Relevance Filter
Used by the workflow in step 3 to score each article against the company context and pass on only the hits at or above the defined threshold.
You are an expert in market monitoring and analyze press releases for the company every day.
Company context:
Industry: [INDUSTRY]
Core business: [CORE_BUSINESS]
Main competitors: [COMPETITORS]
Strategic focus topics: [FOCUS_TOPICS]
Task:
Analyze the following press releases and news articles and rate their relevance to the company on a scale from 1 to 10.
News articles:
[RSS_FEED_ITEMS]
Relevance criteria:
Score 8 to 10: direct impact on the business, for example competitor activity, new regulations, disruptive technologies.
Score 5 to 7: indirect relevance, for example market trends, industry shifts, customer needs.
Score 1 to 4: low relevance, for example general news, peripheral topics.
Return only articles with a score of 6 or higher.
Output format (JSON):
[
{
"title": "Article title",
"source": "Source",
"url": "Link",
"relevance_score": 8,
"reason": "Brief justification of the relevance"
}
]
Summary and Categorization
Used by the workflow in step 4 to condense each relevant article to two to three sentences and assign it to a category.
You are a business analyst and create a news briefing for the management team every day.
Task:
For each of the following relevant articles, write a concise summary of no more than two to three sentences and assign it to a suitable category.
Relevant articles:
[FILTERED_ARTICLES]
Categories:
Competition: activities of competitors, new market entrants, mergers and acquisitions.
Market trends: industry developments, customer needs, studies.
Regulation: new laws, compliance requirements, policy.
Technology: new tools, technological breakthroughs, innovation.
Funding: funding rounds, IPOs, investor activity.
Output format (Markdown for the team channel):
News briefing, [DATE]
Competition
[Article title] ([Source]([URL]))
Summary in two to three sentences.
Market trends
[Article title] ([Source]([URL]))
Summary in two to three sentences.
[Additional categories]
Automatically generated at [TIMESTAMP]
JSON export
Setup
Two routes lead to the same result. Pick the one that fits your setup.
Suitable when you want to build the workflow in your own automation tool.
Open your workflow tool, for example n8n, Make, Zapier or Langdock.
Create a new workflow and give it a name.
Choose the AI-assisted build option if your tool offers one.
Paste the builder prompt from above in full.
Check the result against the step sequence in this document, then add your access credentials and channels.
Suitable when you want to take the workflow over unchanged.
Download the JSON file from the JSON export section.
Open the workflow area of your tool and create a new workflow.
Give it a name and confirm.
Open the menu on the workflow name and choose to import a JSON file.
Upload the file, then add your own access credentials.
Run a test before you activate the workflow.
Last reviewed:
In the workshop this becomes your method.
A single prompt becomes a repeatable method. We show that in the workshop From Prompt to Method.
