I can work on a script that crawls/detects which of these files are like this. I looked at the hexadecimal data of one of the examples from the reddit thread and there might be some indicators. I wasnt able to get a video to plat, but

Just from this file I can see that they created/edited these pdfs using something called reportlab.com, edited on 12/23/2025. Theres also some text refering to Kids? "Count 1/ Kids [ 3 0 R ]" This is odd. screenshot

File EFTS0024813.pdf/mp4? (Dataset 8, folder 0005)

nah, i didn't hear anything back

For anyone looking into doing some OSINT work, this is an epic file EFTA00809187

It contains lists of ALL know JE emails, usernames, websites, social medias, etc from that time

Nice work man! I also discovered something yesterday that I think is worth pointing out.

DUPLICATE FILES: Within the datasets, there are often emails, doc scans, etc that are duplicate entries. (Im not talking about multi torrent stitching, but actual duplicate documents within the raw dataset.) **These duplicates mustbe preserved. ** When looking at two copies of the same duplicate file, I found that sometimes the redactions are in different places! This can be used to extract more info later down the road.

nothing, but event the archived pages arent 100% because some of the files were "faked" in the paginated file lists on the DOJ site. it does work well enough though. I did this to recover all the court records and FOIA files

What does this contain? anything new?

I've been thinking a lot about this whole thing. I don't want to be worried or fearful here - we have done nothing wrong! Anything we have archived was provided to us directly by them in the first place. There are whispers all over the internet, random torrents being passed around, conspiracies, etc., but what are we actually doing other than freaking ourselves out (myself at least) and going viral with an endless stream of "OMG LOOK AT THIS FILE" videos/posts.

I vote to remove any of the 'concerning' files and backfill with blank placeholder PDFS with justification, then collect everything we have so far, create file hashes, and put out a clean + stable archive on everything we have so far. a safe indexed archive We wipe away any concerns and can proceed methodically through blood trail of documents, resulting in an obvious and accessible collection of evidence. From there we can actually start organizing to create a tool that can be used to crowd source tagging, timestamping, and parsing the data. I'm a developer and am happy to offer my skillset.

Taking a step back - Its fun to do the "digital sleuth" thing for a while, but then what? We have the files..(mostly).. Great. We all have our own lives, jobs, and families, and taking actual time to dig into this and produce a real solution that can actually make a difference is a pretty big ask. That said, this feels like a moment where we finally can make an actual difference and I think its worth committing to. If any of you are interested in helping beyond archival, please lmk.

I just downloaded matrix, but I'm new to this, so I'm not sure how that all works. Happy to link up via discord, matrix, email, or whatever.

this dude on pastebin posted his filetree in his epstein ubuntu env - i have a high confidence in whatever lives in his DataSet9Complete.zip file haha

same

Have a scraper running on web.archive.org pulling all previously posted Court-Records and FOIA (docs,audio,etc.) from Jan 30th

Holy shit

The entire Court Records and FOIA page is completely gone too! Fuckers!

Does anyone have the OTHER data sets from before? Ive been lasered in on the DS1-DS12 but havent looked at the other documents at all