Skip to content

VerifyLite

by SecondShotAI · remote · no sign-in to list tools · v1.0.0 · published by GitHub user SecondShotAI

Pay-per-call website safety and URL checks for AI agents. USDC on Base via x402, no API keys.

Changes · 5
CompareRegistry entry

6 tools: 5 change things and 1 only read.

Scanned Oct 21, 2026

1 check matched something in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http verifylite https://verifylite.hydratinghealingtouch.workers.dev/mcp

Findings

Matches of fixed checks, with the text they matched.

1 info
  • Infoverify_url

    The description mentions "credentials".

    Matched text

    description: "...parseable, unsupported scheme, embedded credentials, IP literal, or local/metadata hostname..."

    What this check looks for

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

6 tools

Changes things 5Claude Code: ask
  • check_site_safetyWebsite safety check for one URL.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Website safety check for one URL. Passively fetches the site and returns a verdict (safe, caution, unsafe), a 0-100 score, and evidence-backed flags: TLS validity, HTTP-to-HTTPS redirect, redirect chain and offsite redirects, security headers, mixed content, insecure forms, and domain age from the official registry. Useful before visiting, linking to, buying from, or submitting data to an unfamiliar site. Not charged if the host cannot be checked. Paid via x402; see list_prices.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
  • check_site_safety_proWebsite safety check plus a Google Web Risk lookup for malware, phishing, and unwanted software.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Website safety check plus a Google Web Risk lookup for malware, phishing, and unwanted software. Includes the full /safe report: verdict (safe, caution, unsafe), 0-100 score, TLS, redirects, security headers, mixed content, insecure forms, and domain age. A blocklist hit makes the verdict unsafe. Useful for screening links before a payment, login, or download. Not charged if the blocklist cannot be reached. Paid via x402; see list_prices.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
  • check_sites_batchBatch website safety check: the full /safe report for several URLs in one call, priced per site and cheaper than single checks.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Batch website safety check: the full /safe report for several URLs in one call, priced per site and cheaper than single checks. Each result has a verdict (safe, caution, unsafe), a 0-100 score, and evidence-backed flags for TLS, redirects, security headers, mixed content, insecure forms, and domain age, plus a summary count. Useful for screening every link in a search result, email, or document at once. Not charged if no site can be checked. Paid via x402; see list_prices.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
  • transcribe_audioSpeech-to-text for one audio or video file by URL, using Whisper (turbo) on a GPU.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Speech-to-text for one audio or video file by URL, using Whisper (turbo) on a GPU. Returns the full transcript, timed segments, ready-made SRT and WebVTT captions, and the detected language; optional translation to English. Files up to 25 MB (about 25 minutes of typical MP3). Useful for captioning videos, indexing podcasts, or reading meetings. Not charged if the file can't be fetched or transcribed. Paid via x402; see list_prices.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
  • verify_urlURL syntax validator.1 finding

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Info

      The description mentions "credentials".

      Matched text

      description: "...parseable, unsupported scheme, embedded credentials, IP literal, or local/metadata hostname..."

      What this check looks for
    Description, as the server wrote it
    URL syntax validator. Takes one absolute HTTP or HTTPS URL and returns its normalized form, or a specific error code: unparseable, unsupported scheme, embedded credentials, IP literal, or local/metadata hostname. Never contacts the destination, so it is fast and side-effect free. Useful for cleaning URLs from user input, scraped pages, or LLM output before fetching them. Paid via x402; see list_prices.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: true
Only reads 1Claude Code: allow
  • list_pricesLists every VerifyLite tool and its current price.No findings

    Why it is in this group

    • The name "list_prices" says "list": a word for reading.
    Description, as the server wrote it
    Lists every VerifyLite tool and its current price. Free.

    Annotations the server declares

    • readOnlyHint: true
    • openWorldHint: false

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Install

Pinned to version 1.0.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http verifylite https://verifylite.hydratinghealingtouch.workers.dev/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__verifylite__list_prices"
    ],
    "ask": [
      "mcp__verifylite__verify_url",
      "mcp__verifylite__check_site_safety",
      "mcp__verifylite__check_site_safety_pro",
      "mcp__verifylite__check_sites_batch",
      "mcp__verifylite__transcribe_audio"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. It assumes the server is installed as verifylite. Review it before you use it.