news verifier

    1

    A checker-led workflow for news verification, URL deduplication, and 30-word summary enforcement.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    news verifier

    Example session with this skill installed

    Verify these tech headlines and summarize them in under 30 words:

    1. https://example.com/ai-update?utm_source=rss "New AI Model Released"
    2. https://example.com/ai-update "AI Model v2 News"
    • Read your context and instructions
    • Compiled the news verifier

    I have processed the candidate items. The second item was flagged as a duplicate of the first due to URL normalization (stripping tracking parameters). I verified the first item against the provided source link.

    News Verification Report

    ItemStatusResultSummary
    New AI Model ReleasedVerifiedPassed all 6 checksA new large language model has been released, focusing on improved efficiency and reduced latency for edge devices.
    AI Model v2 NewsDroppedDuplicate URLN/A

    Technical Log

    • source_opens: true
    • headline_matches: true
    • not_rumour: true
    • not_promotion: true
    • own_words_summary: true (21 words)
    • not_duplicate: false (normalized to https://example.com/ai-update)

    Next steps

    1. Run bin/news-verifier dedupe to merge this verified item into your tech.json feed.
    2. Review the rolling cap; this addition will remove the oldest item if the feed exceeds 50 entries.
    3. Publish the updated JSON to your production headline component.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Verify headlines like a desk: check each one against its article page or an official listing, drop anything unverifiable, rumour, promotion or duplicate, and write each summary in your own words (max 30 words). Includes checklists, templates and small offline helpers for dedupe, summary-length checks and rolling-cap enforcement. Checker-led — nothing is fetched or verified automatically.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Workflow Automation.

    Requires human judgement for factual checks. No automated scraping or news API access included. URL deduplication is based on normalized links only, not semantic content.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 4 days ago

    • Free with an Agensi account

    Software
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Environment Variables
    Listed4 days ago

    What's inside

    Frequently Asked Questions

    No, this skill is designed for human-in-the-loop verification. It provides the scripts and structured workflow to enforce word counts and deduplication, but a human must provide the final verdict on factual accuracy.

    The skill includes a dedicated Python script that deterministically counts words for every entry. This eliminates 'prompt drift' where standard LLMs often fail to adhere to strict character or word limits.

    You receive the complete Python-based toolset, standardized JSON templates for news processing, and the structured verification checklist. It includes everything needed to deduplicate URLs and manage feed caps locally.

    The skill uses standard Python (3.10+) and operates entirely on local JSON files. It does not require any external API keys, scrapers, or cloud subscriptions, making it highly secure and private.

    The URL normalization logic automatically strips tracking parameters like UTM and GCLID. This allows the skill to identify and remove redundant links that would otherwise bypass a simple duplicate check.

    New in Workflow Automation

    Free