Ransomware crews pick up the Windows Task Host bug Microsoft fixed in November

Operators suspected of Cl0p affiliation are chaining a pre-authentication information disclosure in the FlexPLM WSDL endpoint to a server-side flaw in the PTC Windchill login servlet, tracked as CVE-2026-12569, to reach unauthenticated remote code execution on internet-exposed product lifecycle management systems, according to an advisory from Hive Pro ^[1][2]. Neither credentials nor user interaction are required, and no encryption stage has been reported, which means the controls most manufacturers bought to catch mass file modification will stay quiet through a complete compromise ^[3][7].

After exploitation, the operators write hex-named JSP web shells into the Windchill login directory, which gives them command execution on the application server ^[4]. Filesystem enumeration follows, with engineering and design data staged for extortion; Hive Pro says the exfiltration channel is unspecified in current reporting ^[5]. Confirmed victim sectors are manufacturing, automotive, aerospace and retail, where the PLM platform holds the drawings and specifications that define competitive position ^[6].

Scoping this by version age will not work. Both products are affected across current version lines, so patch status has to be confirmed per build ^[16]. The vendor advisory lists all builds at or below 11.0 M030 as affected, plus individually enumerated 11.1, 11.2, 12.x and 13.x builds, and Hive Pro directs defenders to check each instance against the exact version table in PTC advisory CS473270 ^[17].

Extortion messaging referencing a serious Windchill PDMLink data leak was first observed on 20 July, sent to hundreds of users inside each affected organisation from randomly compromised accounts and carrying the group's latest contact details ^[8]. Patches have been available since 17 June 2026 ^[10], so the first extortion wave landed 33 days after a fix existed ^[19]. Attribution remains qualified: the actor is unconfirmed, the Cl0p association rests on tradecraft consistent with prior campaigns against enterprise applications, and branded extortion mail and leak-site infrastructure establish brand usage rather than identity ^[9].

The patch date is the wrong starting point for a hunt. Hive Pro notes that exploitation is assessed by one source as likely having begun in early June, before disclosure, though no published indicator predates 18 June ^[11]. That puts at least two weeks of exposure ahead of the patch ^[20], which is why the advisory recommends retrospective web shell hunting back to early June alongside removing these systems from direct internet exposure ^[12]. For detection, the advisory argues the higher-fidelity signal is the Windchill login path itself, because legitimate traffic does not POST there at all, whereas web shell filenames change between deployments ^[13].

As of 28 July, no victims of this campaign had been listed and no credit publicly claimed ^[14], eight days after the extortion mail began ^[21]. Hive Pro assesses that silence as characteristic rather than reassuring: the pattern across prior file-transfer and ERP campaigns is exploit, exfiltrate, extort privately, then mass-publish, making a naming wave in the August to September window plausible and pre-staging of victim tracking worthwhile ^[15].

What to watch: whether any published indicator moves earlier than 18 June, which would confirm the pre-disclosure window rather than infer it ^[11]; whether a leak site naming wave arrives in the August to September window ^[15]; and whether POST traffic to the Windchill login path proves as clean a signal in production as the advisory claims, since that assumption is what the whole detection approach rests on ^[13].