#exploitarium
Anonymous researcher drops 0-day 'exploitarium' repo
Anonymous researcher drops 0-day 'exploitarium' repo
At least two vulnerabilities are already under attack
www.theregister.com
July 2, 2026 at 4:34 AM
Anonymous researcher drops 0-day 'exploitarium' repo
Anonymous researcher drops 0-day 'exploitarium' repo
At least two vulnerabilities are already under attack
www.theregister.com
June 29, 2026 at 8:30 PM
GitHub’s Exploitarium Drops Ready-to-Run Code for Critical Flaws in libssh2, Gitea and More An anonymous researcher has consolidated dozens of functional exploit PoCs into the Exploitarium reposi...

#CybersecurityUpdate #AI #agent #security #exploitarium […]

[Original post on webpronews.com]
June 28, 2026 at 10:39 AM
What happens when AI uncovers zero-day exploits before developers can fix them?

On Security Boulevard, Tom Hollingsworth & Fernando Montenegro discuss the "Exploitarium" GitHub code drop, where AI exposed zero-day vulnerabilities in critical open-source software. #SecurityBlvd #Cybersecurity
July 24, 2026 at 5:49 PM
Anonymous researcher drops 0-day 'exploitarium' repo
Anonymous researcher drops 0-day 'exploitarium' repo
www.theregister.com
June 29, 2026 at 9:08 PM
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41 🎙️ buff.ly/7N1C5Aw

@NetworkingNerd.net @SFMontenegro.bsky.social #SecurityBlvd #Cybersecurity #ZeroDay #GitHub #OpenSource #AI #AISecurity
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41
What happens when an anonymous researcher uses AI to unleash a hidden wave of zero-day exploits without giving maintainers any warning? In this episode of the Security Boulevard Podcast, hosts Tom…
buff.ly
July 21, 2026 at 10:56 PM
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41 🎙️ buff.ly/7N1C5Aw

@NetworkingNerd.net @SFMontenegro.bsky.social #SecurityBlvd #Cybersecurity #ZeroDay #GitHub #OpenSource #AI #AISecurity
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41
What happens when an anonymous researcher uses AI to unleash a hidden wave of zero-day exploits without giving maintainers any warning? In this episode of the Security Boulevard Podcast, hosts Tom…
buff.ly
July 25, 2026 at 8:57 PM
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41 🎙️ youtu.be/Pf3HZ7ipSeQ

@NetworkingNerd.net @SFMontenegro.bsky.social #SecurityBlvd #Cybersecurity #ZeroDay #GitHub #OpenSource #AI #AISecurity
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41
What happens when an anonymous researcher uses AI to unleash a hidden wave of zero-day exploits without giving maintainers any warning? In this episode of the Security Boulevard Podcast, hosts Tom…
buff.ly
July 23, 2026 at 9:51 PM
Researcher Behind 'Exploitarium' Explains Release of Undisclosed Zero

arc-codex.com
July 3, 2026 at 7:43 AM
Anonymous researcher drops 0-day 'exploitarium' repo
Anonymous researcher drops 0-day 'exploitarium' repo
At least two vulnerabilities are already under attack
www.theregister.com
June 30, 2026 at 8:32 PM
Anonymous researcher drops 0-day 'exploitarium' repo

www.theregister.com/security/202...

#Kyberturvallisuus #ThreatIntel #Haavoittuvuus
Anonymous researcher drops 0-day 'exploitarium' repo
At least two vulnerabilities are already under attack
www.theregister.com
July 9, 2026 at 10:55 AM
Von PHP bis RustDesk: Der Unbekannte hat mit KI-Unterstützung Sicherheitslücken in allerlei quelloffener Software entdeckt und publik gemacht [--] als Geschenk. #Security
Exploitarium: Anonymer Sicherheitsforscher veröffentlicht zwei Dutzend Zero-Days
Von PHP bis RustDesk: Der Unbekannte hat mit KI-Unterstützung Sicherheitslücken in allerlei quelloffener Software entdeckt und publik gemacht [--] als Geschenk.
www.heise.de
July 1, 2026 at 8:03 AM
Somebody posted a bunch of exploits on GitHub that claim to be 0days.
https://github.com/bikini/exploitarium

Let's look at the first one: A 7-Zip MotW bypass.

**Problem #1** : 7-Zip **DOESN'T EVEN WRITE MOTW FOR EXTRACTED FILES BY DEFAULT**. This doesn't […]

[Original post on infosec.exchange]
June 27, 2026 at 3:34 AM
Anonymous GitHub account mass-dropping undisclosed 0-days
https://github.com/bikini/exploitarium
[comments] [278 points]
June 27, 2026 at 6:03 PM
9. https://github.com/magnitudedev/magnitude - TypeScript, 395 stars today
10. https://github.com/bikini/exploitarium - Python, 68 stars today
11. https://github.com/bannedbook/fanqiang - Kotlin, 735 stars today
12. https://github.com/debpalash/VoiceStudio - Python, 1345 stars today
13 […]
Original post on backend.newsmast.org
backend.newsmast.org
September 4, 2026 at 3:47 PM
🕳️ ALPHA · score 9/10

Anonymous GitHub account mass-dropping undisclosed 0-days

https://github.com/bikini/exploitarium
June 27, 2026 at 4:27 PM
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41 🎙️ buff.ly/7N1C5Aw

@NetworkingNerd.net @SFMontenegro.bsky.social #SecurityBlvd #Cybersecurity #ZeroDay #GitHub #OpenSource #AI #AISecurity
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41
What happens when an anonymous researcher uses AI to unleash a hidden wave of zero-day exploits without giving maintainers any warning? In this episode of the Security Boulevard Podcast, hosts Tom…
buff.ly
July 24, 2026 at 5:49 PM
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41 🎙️ buff.ly/7N1C5Aw

@NetworkingNerd.net @SFMontenegro.bsky.social #SecurityBlvd #Cybersecurity #ZeroDay #GitHub #OpenSource #AI #AISecurity
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41
What happens when an anonymous researcher uses AI to unleash a hidden wave of zero-day exploits without giving maintainers any warning? In this episode of the Security Boulevard Podcast, hosts Tom…
buff.ly
July 21, 2026 at 8:57 PM
Exploitariumによる大規模ゼロデイアーカイブ、35個のオープンソースプロジェクトを標的とする204個のPoCファイルを公開

「bikini」として知られるGitHubアカウントが、Exploitariumという名称で活動し、353535個のオープンソースプロジェクトを標的とした脆弱性概念実証(PoC)ファイルの大規模かつ拡大を続けるアーカイブを公開しました。 このリポジトリには204204204件の追跡対象ファイルが含まれており、これまで...
Exploitariumによる大規模ゼロデイアーカイブ、35個のオープンソースプロジェクトを標的とする204個のPoCファイルを公開
「bikini」として知られるGitHubアカウントが、Exploitariumという名称で活動し、353535個のオープンソースプロジェクトを標的とした脆弱性概念実証(PoC)ファイルの大規模かつ拡大を続けるアーカイブを公開しました。 このリポジトリには204204204件の追跡対象ファイルが含まれており、これまで
blackhatnews.tokyo
July 22, 2026 at 7:20 AM
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41 🎙️ buff.ly/7N1C5Aw

@NetworkingNerd.net @SFMontenegro.bsky.social #SecurityBlvd #Cybersecurity #ZeroDay #GitHub #OpenSource #AI #AISecurity
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41
What happens when an anonymous researcher uses AI to unleash a hidden wave of zero-day exploits without giving maintainers any warning? In this episode of the Security Boulevard Podcast, hosts Tom…
buff.ly
July 25, 2026 at 10:56 PM
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41 🎙️ buff.ly/7N1C5Aw

@NetworkingNerd.net @SFMontenegro.bsky.social #SecurityBlvd #Cybersecurity #ZeroDay #GitHub #OpenSource #AI #AISecurity
AI-Powered Zero-Days: The "Exploitarium" Open Source Chaos | Security Boulevard Podcast Episode 41
What happens when an anonymous researcher uses AI to unleash a hidden wave of zero-day exploits without giving maintainers any warning? In this episode of the Security Boulevard Podcast, hosts Tom…
buff.ly
July 21, 2026 at 5:49 PM
Anonymous Researcher Dumps 204 0-Day Exploit Files Before Vendors Can Patch Them
Anonymous Researcher Dumps 204 0-Day Exploit Files Before Vendors Can Patch Them
An anonymous GitHub user has quietly assembled one of the most disruptive exploit collections of the year, dropping 204 zero‑day proof‑of‑concept files for dozens of open‑source projects before vendors had a chance to patch them. The archive, hosted under the name “exploitarium” by a researcher using the handle “bikini,” turned coordinated disclosure on its head by publishing exploit code first and leaving vendors and defenders to catch up in public. What initially looked like a short‑lived spike in attention has instead evolved into a growing repository that continues to add new vulnerable projects week after week. The exploitarium archive started to take shape in late June 2026, with dated entries appearing from June 23 and a first large public batch drawing heavy traffic around June 27. As issues and pull requests piled up, the repository’s README made clear that none of the findings had been reported to affected vendors ahead of time, and explicitly encouraged others to claim credit by disclosing them independently. Analysts from LevelBlue SpiderLabs noted that this approach put exploit code in front of attackers and defenders simultaneously, erasing the protective window that coordinated disclosures are designed to provide. Since that initial burst of attention, the project has continued to expand while most public coverage moved on. LevelBlue said in a report shared with Cyber Security News (CSN) that the new folders have landed at a steady pace of roughly two to three per week, including recent additions against widely deployed platforms such as PostgreSQL, Redis, Nextcloud, and Discourse in the first week of July alone. Illustration of the pre-fix size computation (Source – LevelBlue) Behind the headline number of “130 PoCs” cited in early reporting, SpiderLabs’ count now shows 35 tracked project folders with 204 files, reflecting just how much research has been folded into the archive after the first stories were published. Illustration of the upstream fix (commit 97acf3d) (Source – LevelBlue) The attack surface exposed by these PoCs spans far more than a single stack or vendor. Native C and C++ codebases, kernel drivers, web application logic, remote‑access tools like AnyDesk and RustDesk, and core infrastructure such as Nmap and curl all appear in the inventory. This breadth turns the incident into a supply chain problem as much as a vulnerability story, since many organizations may inherit exploitable libraries as transitive dependencies without realizing they are there. Anonymous Researcher Dumps 204 0-Day Exploit Files The anonymous drop closely matches a pattern that LevelBlue had already seen this year in the Nightmare‑Eclipse campaign, but with a wider and more systematic reach across open‑source ecosystems. Whereas Nightmare‑Eclipse zeroed in on a single vendor and Windows internals, exploitarium pushes mass disclosure into developer tools, cloud‑ready services, container platforms, and community forums. For readers who followed earlier coverage of Windows Defender exploitation, the RoguePlanet articles on  new Windows Defender 0‑Day exploit RoguePlanet  and RoguePlanet Defender patch claim may leak data show how quickly uncoordinated drops can translate into complex operational risk. External contributor adds new vulnerability research via Pull Request #4 (Source – LevelBlue) Within exploitarium, not every folder carries the same weight, and LevelBlue’s assessment stresses that defenders should avoid treating the entire archive as uniformly critical. Many entries look like low‑impact artifacts from automated fuzzing runs, while others document high‑severity, pre‑authentication flaws in widely embedded components. A standout example is CVE‑2026‑55200 in libssh2, an out‑of‑bounds write in the ssh2_transport_read function that can be triggered before any SSH authentication completes, potentially enabling remote code execution against vulnerable clients. Cyber Security News (CSN) previously covered this bug in detail in its piece on the  critical libssh2 vulnerability allows attackers  and a follow‑up on the  PoC exploit released libssh2 RCE vulnerability , showing how a single entry from the archive can ripple quickly into real‑world attack scenarios. The libssh2 case also illustrates the messy overlap between coordinated and uncoordinated disclosure. While exploitarium included a libssh2 PoC from day one, the official CVE credit went to researcher Tristan Madani through VulnCheck’s normal reporting channels, meaning the same underlying defect surfaced through both responsible and mass disclosure paths. For defenders, that nuance matters less than the practical takeaway: track the upstream fix, migrate to a version that incorporates commit 97acf3d, and treat the vulnerability’s broad transitive footprint as a priority regardless of whether exploitation has been confirmed. From solo archive to shared risk Over time, exploitarium has shifted from a personal archive to a lightly moderated clearinghouse that accepts outside research. LevelBlue points to contributions such as Pull Request 4 from the user “Unrealisedd,” which introduced new Windows kernel driver issues in OpenVPN’s ovpn‑dco‑win component, complete with suggested remediations. The maintainer also retroactively credited external work on objdump, acknowledging that another researcher had published a more complete PoC separately before the finding was folded into the archive. For organizations that worry about provenance, this blend of careful Git‑tree verification and informal contributor vetting means exploitarium now amplifies unvetted code with the implicit credibility of its growing collection. The README details a rigorous provenance verification process (Source – LevelBlue) LevelBlue’s guidance to security and detection teams centers on pragmatic triage under pressure. Patch first and rank items by reachability and blast radius, not by media attention or proof‑of‑concept novelty alone. They recommend running software composition analysis across build pipelines, container images, and vendored libraries to find hidden libssh2 dependencies and other embedded components touched by the archive. In parallel, they highlight community efforts like the Exploitarium‑Detections project, which publishes KQL rules for Microsoft Sentinel and Defender XDR to spot activity associated with the disclosed vulnerabilities. For organizations already tracking recent supply chain incidents such as the  Mastra npm packages compromised attack , exploitarium should be read as part of the same broader trend: research repositories that can instantly shift theoretical zero‑days into accessible attack kits. LevelBlue’s report underscores that exploitarium is “not a discrete incident with a start and end date” but an ongoing infrastructure likely to keep generating both genuine findings and triage overhead for as long as it is maintained. The maintainer has already paired the archive with a general‑purpose code‑obfuscation toolkit, underlining that this is an active research pipeline rather than a one‑off disclosure stunt. For executive teams, the recommended framing is closer to a long‑tail supply chain exposure than a single vulnerability, demanding sustained monitoring, rapid patch adoption, and continual reassessment of which open‑source components are quietly bundled into critical workloads. Indicators of Compromise (IoCs):- Type Indicator Description Project folder libssh2-cve-2026-55200-poc Proof‑of‑concept archive for CVE‑2026‑55200 affecting libssh2. Project folder objdump-dlx-calc-poc Large objdump exploit collection with 41 tracked files. Project folder rustdesk-session-permission-pocs RustDesk session permission exploit set with 17 tracked files. Project folder ovpn-dco-win (Windows kernel driver) OpenVPN Windows kernel‑mode driver vulnerabilities added via PR 4. Project folder nmap-ipv6-extlen-wrap-poc Nmap IPv6 extension length wrap exploit project folder. Project folder postgres-ri-owner-switched-cast-poc PostgreSQL referential integrity owner switch exploit folder. Project folder redis-vset-duplicate-hnsw-id-rce-poc Redis RCE via vset duplicate HNSW ID exploit project. Project folder nextcloud-federated-share-bearer-token-poc Nextcloud federated share bearer token abuse exploit folder. Project folder discourse-scoped-api-key-preauth-bypass Discourse scoped API key pre‑authentication bypass folder. Project folder firefox-smartwindow-private-url-exfil-poc Firefox Smart Window private URL data exfiltration exploit project. Note:   IP addresses and domains are intentionally defanged (e.g.,  [.] ) to prevent accidental resolution or hyperlinking. Re-fang only within controlled threat intelligence platforms such as MISP, VirusTotal, or your SIEM . ! ALERT: 20+ government sites delivered malware to businesses and citizens.  See full attack research  to check your own exposuremalware to businesses an The post Anonymous Researcher Dumps 204 0-Day Exploit Files Before Vendors Can Patch Them appeared first on Cyber Security News .
cybersecuritynews.com
July 22, 2026 at 9:53 AM
What happens when AI uncovers zero-day exploits before developers can fix them?

On Security Boulevard, Tom Hollingsworth & Fernando Montenegro discuss the "Exploitarium" GitHub code drop, where AI exposed zero-day vulnerabilities in critical open-source software. #SecurityBlvd #Cybersecurity
July 21, 2026 at 5:49 PM
🧵Un investigador anònim anomenat bikini ha publicat a GitHub un repositori amb PoCs d'exploits per a més de 15 projectes open source populars, anomenat exploitarium. En alguns casos sense avisar cap mantenidor abans de publicar. Dos ja tenen CVE crític confirmat.
July 1, 2026 at 7:44 AM