In my phishing email, I want to render the HTML of the email in an artifact so the analyst can see the email as the user would.
Currently I convert it to Markdown and that only works slightly.
I’ve tried using some markdown escapes, but they just render the raw HTML (which I have in the next Artifact for additional review)
Any idea how I can render these or if Artifacts will ever support other input types?
1 Like
Have you tried using the HTML plugin to convert the HTML to markdown?
1 Like
That is the one I originally used, it doesn’t do a good job when emails have tables and other formating
1 Like
I don’t know if this trick will work for your specific example, but it’s one I’ve used before. It’s something I discovered by accident when I was setting up my phishing workflow. Since the screenshot plugin spits out Base64, I started thinking of what I could do with it. In my case I send the encoded image to the GPT-4 API and use it to perform a comparison of the screens behind any out links to our CAS and MS portals (images also encoded in base64) and the result is the probability of an out link leading to a counterfeit authentication page. After doing some training of the AI, it’s nearly spot on in detecting fake logins. I was thinking you might be able to try converting the HTML to base64 to store in the artifact, then decode it later for viewing. I’ve never tested if there is a character limit for artifacts, so my only concern would be if the base64 string is so long it exceeds the maximum artifact length.
I got curious and started experimenting. What i don’t know is how you’re viewng the results of your workflow. I send the artifacts and relevant loglines to a Google sheet and use appscript for post processing to create the graphs for users affected. Number who read the retracted message, etc and render the final report. I could make it work in my post processing but don’t how i could do the same thing purely in ICONN.