---
name: linksafe.melis.ai
description: linksafe.melis.ai provides a single URL safety-checking service that evaluates a given URL for phishing, malware, and suspicious content, returning a risk verdict and associated metadata. It is a focused, single-purpose endpoint with no bulk or file-scanning capabilities.
host: linksafe.melis.ai
---

# linksafe.melis.ai

linksafe.melis.ai is a URL threat-assessment host serving agents and users who need to verify whether a link is safe before visiting, sharing, or processing it. It targets phishing and malware detection at the individual URL level, with no broader security scanning scope.

## When to use this host

Use this host when an agent needs to assess the safety of a specific URL before acting on it — for example, before redirecting a user, embedding a link, or processing a URL from untrusted input. Do not use it for scanning file contents, email attachments, or IP addresses directly. It does not support batch/bulk URL checks in a single request; if you need to evaluate many URLs at once, you will need to call the skill once per URL or seek a bulk-scanning alternative. For broader threat intelligence (domain reputation, WHOIS, IP analysis), look for a dedicated threat-intelligence host rather than this one.

## Capabilities

### URL Safety Verification

Evaluates a single URL against known phishing, malware, and suspicious-content signals, returning a risk verdict and metadata to inform downstream decisions.

- **`verify-url-safety`** — Checks a URL for phishing, malware, and suspicious content, returning a risk verdict and metadata.

## Skill reference

### `verify-url-safety`

**LinkSafe Verify** — Checks a URL for phishing, malware, and suspicious content, returning a risk verdict and metadata.

*Use when:* Use when an agent or user needs to assess whether a URL is safe before visiting, sharing, or processing it — particularly to detect phishing or malware links.

*Not for:* Do not use for scanning file contents or email attachments; this endpoint only evaluates URLs. Not suitable for bulk URL scanning in a single request.

**Inputs:**

- `url` (string, required) — The URL to check for safety. Must be a valid, well-formed URL string.

**Returns:** Returns a JSON object with a risk verdict and metadata on success, or an error string describing the failure reason on invalid input.

**Example:** `{"url": "https://example.com"}`

---