Catch Fake Links Before You Publish: LLM Hallucination Check

    1

    Unlike AEO audits that score AI visibility — catch fabricated URLs, ASINs, and citations in LLM drafts before you publish. Offline flags + optional live HEAD checks. Stdlib script included.

    $7

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Catch Fake Links Before You Publish: LLM Hallucination Check

    Example session with this skill installed

    Verify the links in this product roundup draft for the new smart home collection. Use the --live flag to check for dead Amazon links.

    • Read your context and instructions
    • Compiled the catch fake links

    I have processed the product roundup draft. The script extracted 12 unique URLs and performed both offline pattern matching and live HEAD/GET verification.

    Link QA Report

    URLContextVerdictFlags/StatusNotes
    https://amazon.com/dp/B00EXAMPLE1"Smart Hub Pro"FAILASIN_SHAPE_MISMATCHLink follows invalid ASIN format; likely hallucinated.
    https://homelife.test/setup/todo"Setup Guide"FAILPLACEHOLDER_PATHContains '/todo' path marker common in AI drafts.
    https://best-tech.com/review-2024"Expert Review"WARN404_NOT_FOUNDServer returned 404. Live check failed.
    https://standard-io.com/docs"Documentation"PASS200_OKValid resolution to final destination.
    https://partner-link.net/ref"Buy Now"WARNDOMAIN_MISMATCHAnchor text domain does not match href domain.

    Next steps

    • Replace the Smart Hub Pro ASIN with the verified ID from your Amazon Associates dashboard.
    • Update the 'Setup Guide' link to the final production URL.
    • Remove or fix the 'Expert Review' citation as the destination page no longer exists.

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

    About this skill

    Unlike AEO/citation audits that score pages for AI visibility — this catches fabricated URLs, ASINs, and citations in LLM drafts before you publish (common failure: ~80%+ fake product links in unchecked AI drafts).

    Sample output (offline dry-run): urls=15 | PASS=2 | WARN=3 | FAIL=10 flags: suspicious_tld, asin_shape_mismatch, placeholder_domain, fake_path_marker, bare_ip_host, near_duplicate, link_text_domain_mismatch, javascript_scheme

    CSV columns: url, source_context, offline_flags, live_status, final_url, verdict (PASS|WARN|FAIL), notes

    What you get

    • SKILL.md workflow for Claude Code / Cursor / Codex
    • scripts/link_verify.py (Python stdlib only; --live optional HEAD/GET on URLs you supply)
    • Example draft + deterministic sample CSV
    • templates/link-qa-report.md

    Works in Claude Code, Cursor and Codex · tested on sample drafts · v1.0.0 Oct 2026 · questions answered within 24h

    Pairs with: AEO Audit, SEO Content Brief, Source-Backed Research Brief.

    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 any payment.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 13 days ago

    Listed13 days ago

    What's inside

    Frequently Asked Questions