A PowerShell script that takes a Discourse topic URL and prints all posts containing external links. This is useful for identifying spam posts that hide links in subtle places, such as within commas or periods. If a post seems suspicious, run this script with the Discourse topic URL as a parameter to display all external links found in the posts.

Check-Discourse-Topic-Links.ps1

<#
.SYNOPSIS
Takes a discourse topic url and prints all posts that contain external links. Useful for identifying spam posts that hide links in subtle places, such as within commas or periods.

.DESCRIPTION
The Check-Discourse-Topic-Links script takes a URL and uses the Invoke-RestMethod to check each post for the presence of a link_counts field. If the field exists, the script checks whether its internal property is set to false. If so, it prints the post number, the username of the post's author, the external link and the link text.

.PARAMETER discourseTopic
The discourse topic url to check. Note: the url must begin with http:// or https://

.EXAMPLE
.\Check-Discourse-Topic-Links.ps1  https://discussions.unity.com/t/is-it-a-good-idea-to-put-my-game-on-itch-io-first/1657629
#>

[CmdletBinding()]
param
(
	[Parameter(Mandatory=$True)]
	[string]$discourseTopic
)

$pageNumber = 0
Write-Host "`n"

while($True)
{
	$pageNumber++
	$currentPage = $discourseTopic + ".json?page=" + $pageNumber
	
	try
	{
		Write-Verbose "Getting all posts from $discourseTopic page $pageNumber"
		$pageData = (Invoke-RestMethod -UserAgent ([Microsoft.PowerShell.Commands.PSUserAgent]::Chrome) -Method GET -Uri $currentPage)
	}
	catch
	{
		exit
	}

	$allPosts = $pageData.post_stream.posts

	foreach($post in $allPosts)
	{
		Write-Verbose "Checking post: $($post.post_number) by $($post.username) `n"
		if($post.link_counts)
		{
			Write-Verbose "Links found in post number $($post.post_number) `n"
			foreach($link in $post.link_counts)
			{
				Write-Verbose "Link: $link `n"
				if($link.internal -eq $False)
				{
					Write-Output "External link found in post number: $($post.post_number) by $($post.username)"
					Write-Output "$($link.url)"
					
					$escapedLink = [regex]::Escape($($link.url))
					$pattern = "<a\s+[^>]*href=`"$escapedLink`"[^>]*>(.*?)</a>"
					
					if($post.cooked -match $pattern)
					{
						$linkText = $matches[1]
						Write-Output "Link Text: $linkText `n"
					}
					else
					{
						Write-Output "No link text found `n"
					}		
					
					Write-Verbose "Post body: $($post.cooked) `n"
				}
			}
		}
	}
}

CheckDiscourseTopicLinks.cs

// A C# script with the same functionality as the above powershell script

using System.Text.Json;
using System.Text.RegularExpressions;

if (args.Length != 1)
{
    Console.WriteLine("Usage: Program <DiscourseTopicUrl>");
    return;
}

string discourseTopic = args[0];
int pageNumber = 0;
HttpClient client = new();
client.DefaultRequestHeaders.UserAgent.ParseAdd("Mozilla/5.0 (compatible; Chrome)");

Console.WriteLine();

while (true)
{
    pageNumber++;
    string currentPage = $"{discourseTopic}.json?page={pageNumber}";

    try
    {
        string response = await client.GetStringAsync(currentPage);
        using JsonDocument document = JsonDocument.Parse(response);
        JsonElement allPosts = document.RootElement.GetProperty("post_stream").GetProperty("posts");

        foreach (var post in allPosts.EnumerateArray())
        {
            if (!post.TryGetProperty("link_counts", out var allLinks) ||
                allLinks.ValueKind != JsonValueKind.Array) continue;

            foreach (var link in allLinks.EnumerateArray())
            {
                ProcessLink(link, post);
            }
        }
    }
    catch (HttpRequestException)
    {
        break;
    }
}

return;

static void ProcessLink(JsonElement link, JsonElement post)
{
    if (!link.TryGetProperty("internal", out var internalLink) ||
        internalLink.ValueKind != JsonValueKind.False) return;

    int postNumber = post.GetProperty("post_number").GetInt32();

    string? url = link.GetProperty("url").GetString();
    if (url == null)
        throw new NullReferenceException($"link: {link} url property from the parsed json post: {post} is null!");

    string? username = post.GetProperty("username").GetString();
    if (username == null)
        throw new NullReferenceException($"link: {link} username property from the parsed json post: {post} is null!");

    string? postBody = post.GetProperty("cooked").GetString();
    if (postBody == null)
        throw new NullReferenceException($"link: {link} cooked property from the parsed json post: {post} is null!");

    string escapedLink = Regex.Escape(url);
    string pattern = $"""<a\s+[^>]*href=\"{escapedLink}\"[^>]*>(.*?)</a>""";

    Console.WriteLine($"External link found in post number: {postNumber} by {username}");
    Console.WriteLine(url);

    Match match = Regex.Match(postBody, pattern);
    if (match.Success)
    {
        string linkText = match.Groups[1].Value;
        Console.WriteLine($"Link Text: {linkText}\n");
    }
    else
    {
        Console.WriteLine("No link text found\n");
    }
}