Loading vulnerability…
NLTK contains an argument-injection flaw in per-call JVM option handling used by GenericStanfordParser, StanfordTagger, StanfordTokenizer, and StanfordSegmenter. User-supplied java_options passed to nltk.internals.java() are not validated when supplied through the per-call options parameter, bypassing JVM-option validation. An attacker who controls this value can introduce arbitrary JVM arguments, including options that load Java or native agents, enable remote debugging, or cause additional argument files to be processed.
Mallory correlates every CVE against your assets, your vendors, and active adversary campaigns. Know which vulnerabilities matter for you, not just which ones are loud.
What it means. What to do now. Patch path, mitigations, and the assume-compromise checklist.
What an attacker gets, and what they’ve been doing with it.
If you can’t patch tonight, do this now.
Patch, then assume compromise.
No public exploits tracked yet. Mallory keeps watching.
No public exploit code observed for this vulnerability.
1 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.
Query your assets running an affected version, and investigate the blast radius.
Every observed campaign linking this CVE to a named adversary.
Malware families riding this exploit, with evidence and IOCs.
YARA, Sigma, Snort, and vendor rules, auto-deployed to your SIEM.
Cross-references every affected SKU, including bundled OEM variants.
Community discussion across Reddit, Mastodon, and other social sources.