Situational Awareness Terminal
◈ Source Credibility Index
1. BLUF (Bottom Line Up Front)
An unidentified actor began compromising ConnectWise ScreenConnect instances used by MSPs on November 11, 2026. The actor used them to gain unauthorized access and to deploy the SCKit worm and unauthorized remote-access clients on customer-managed endpoints, leveraging ScreenConnect's remote-support platform to establish persistence and distribute payloads to technician workstations and managed devices. MSPs and their downstream customers are the affected population, and the persistent unauthorized remote-access clients are the realized risk.
2. Key Judgments — SCKit ScreenConnect MSP Campaign
- The realized impact is mass unauthorized remote access, not worm propagation. Huntress reports at least 288 endpoints across 48 organizations compromised within 90 minutes, about one-third in the first 10 minutes. Microsoft says the added clients connect to attacker-controlled ScreenConnect infrastructure and survive termination of the MSP's session.
- The RMM+ empty-access-key flaw is the best-supported entry route but is not confirmed. Huntress "believes" it was the route and found commands executed with no matching user logins, with logs linking them to RMM+ versions 1.0.31 and 1.0.32. Microsoft, citing ConnectWise, says the entry method "remains under investigation."
- The technician-compromise chain (RelayTrap, cookie theft, Shared Toolbox distribution) is documented as design intent. Blackpoint reproduced the exploit against a current client and the patch stopped it. Huntress says the console takeover failed in every recovered host log it examined. Microsoft's own figure caption calls the path "the toolkit's propagation design, not a confirmed sequence of infections."
- Removing a rogue client does not clean a host. Microsoft says persistence (scheduled tasks, a registry Run entry named RelayTrapLiteAlert, files under C:\ProgramData\RelayTrap) and exposed web sessions survive that step.
- Huntress says RelayTrap appears to be LLM-generated, and BleepingComputer notes the source comments. This describes how the code was written. It says nothing about actor identity or capability.
3. Analysis of Competing Hypotheses (ACH)
| Hypothesis | Supporting Evidence | Contradicting Evidence | Evidence Gaps | Probability |
|---|---|---|---|---|
| H-A: One actor abused the RMM+ empty-key flaw to mass-deploy rogue clients and SCKit; the technician-worm stage was attempted but has not succeeded. | Huntress: commands ran across multiple instances without user logins, logs tie them to RMM+ 1.0.31/1.0.32, and a blank key passes the check. ConnectWise disabled five extensions in about 400 partner environments. Blackpoint reproduced the exploit failing; Huntress saw takeover fail in every recovered log. Microsoft's figure caption describes propagation as design. | ConnectWise says the entry method is still under investigation. Microsoft's text says the worm "targets" technician workstations via hijacked sessions without separating intent from observed outcome. | Telemetry from MSPs that did not report. Whether all 48 organizations had RMM+ with a blank key. Actor identity. | 66% |
| H-B: Initial access came through more than one route (other extensions, credential theft, other server flaws), with RMM+ one of several. | ConnectWise disabled five extensions, not one. Microsoft says extension abuse is an investigated route, not a confirmed one. Huntress frames the RMM+ link as belief. | No source reports an alternative route. The log pattern of commands with no login events matches the RMM+ mechanism. ConnectWise said extension presence does not mean compromise. | Per-instance forensic data. ConnectWise findings on the other four extensions. | 17% |
| H-C: The worm stage has worked against some technicians, outside the researchers' telemetry. | Microsoft says it "identified" technician targeting via hijacked sessions. Cookie theft would not need the CVE-2026-84869 path once RelayTrap runs on a technician host. About 400 partner environments were touched by the extension shutdown. | Blackpoint found "no evidence of successful exploitation and no evidence of a bypass." Huntress saw failure in every recovered log. ConnectWise says the fix blocks the spread. | Unpatched technician clients. Browser-session audit logs. Shared Toolbox activity. | 11% |
| H-D (Maskirovka / Strategic Deception): The apparent signal is a deliberate disinformation, fabrication, or denial-and-deception operation designed to shape perception or mask a different course of action. | The campaign uses a legitimate tool and a "c=Exploited<N>" marker. The toolkit has a long, commented, LLM-style code body. Several components appear unfinished or untested, which could mean decoys or noise. | Two independent research teams recovered the loader from attacker infrastructure and decrypted later stages. Microsoft observed the activity separately. Scale (288 endpoints in 90 minutes) is consistent with real operations. | Infrastructure analysis (ns[.]mo7sen[.]dev and the attacker-controlled ScreenConnect servers). Any second-stage activity. | 4% |
ACH Assessment: H-A is best supported because it fits the most evidence and conflicts with none directly. The main tension is between Microsoft's present-tense wording on technician targeting and the researchers' failed-exploitation findings. We read this as differing emphasis (what the toolkit does versus what has been seen working), not a factual conflict. It still keeps H-C open. The two independent source families are Microsoft and BleepingComputer, and the latter relies on Huntress and Blackpoint reports, so the evidence base is narrower than a count of three documents suggests.
4. Key Assumption Check (KAC)
- Critical Assumptions:
(1) Researchers' failed-exploit tests reflect the real client population. If many technicians run clients older than 26.6.5, H-C rises sharply.
(2) The 288-endpoint/48-organization count is a floor, as Huntress says ("at least"). If undercounted, scale and downstream exposure are larger.
(3) Patch effectiveness against the file-transfer path implies no alternative path to technicians. Cookie theft or console takeover by other means is not ruled out by the patch.
(4) Activity described by Microsoft and Huntress belongs to the same campaign. If separate actors are involved, attribution and mitigation change.
(5) The United States location is inferred in the reporting from ConnectWise and Huntress involvement; no source states the geography. If victims are international, regulatory and sector exposure shifts. - Information Gaps: Actor identity and motive are not reported. The confirmed entry route is not reported. There is no data on what was done with the access beyond installing clients (no ransomware, data theft, or lateral movement is described). ConnectWise's evidence on the other four disabled extensions is not given, nor are victim sectors. Closing these requires ConnectWise's investigation outcome, host forensics from affected MSPs, and infrastructure pivots on the attacker-controlled servers.
- Bias & Deception Risks: Vendor framing bias is possible: Microsoft, Huntress, and Blackpoint have commercial incentives to emphasize detection value, and ConnectWise has an incentive to stress that the patch holds and that extension presence does not mean compromise. There is single-origin echo risk because BleepingComputer's account relies on Huntress and Blackpoint. The timeline records no contradictions, but the unresolved intent-versus-outcome wording above is a signal to track. No source indicates deliberate adversary deception. The LLM-generated appearance of the code is a noted characteristic, not evidence of misdirection.
5. Implications and Strategic Risks — ConnectWise ScreenConnect and MSP Supply Chain
The near-term risk is dormant persistent access on customer endpoints, which can be used later by this actor or sold. The patch blocked the worm step this time, but the toolkit's design shows that technician sessions are now a target class. If attackers find another route to technician consoles, one compromised MSP could reach many customers. Disabling extensions by the vendor was an unusually broad action, so further advisories on related extensions are plausible.
Cyber / Information Space — MSP Technician Workstations and ScreenConnect Consoles
Microsoft says removing an unauthorized client leaves persistence and exposed web sessions in place, so incomplete cleanup is a likely source of re-compromise. Cookie theft would bypass credentials and a new MFA challenge, which makes session invalidation a distinct task from patching. The hardcoded c=Exploited<N> marker and the staged URL paths give defenders concrete detection handles.
Economic / Social — Managed Service Provider Customers
Affected MSPs face investigation costs across customer fleets and may have to notify clients. The researchers' advice to avoid interactive ScreenConnect sessions with possibly infected endpoints complicates routine support work. Customer trust in RMM tooling could shift trust toward vendors that publish extension audits and isolation controls.
Security — Third-Party Extension Ecosystem
The RMM+ flaw is a configuration-dependent weakness: a blank key passed authentication when an administrator never set one. Similar default-state weaknesses may exist in other extensions, and ConnectWise disabled five. Adversaries tend to reuse such patterns, so extension APIs that bypass normal sign-in are a likely focus of research and abuse.
6. Recommendations and Outlook
- Immediate Actions (0–30 days): Track whether ConnectWise confirms the entry route and whether the count of 288 endpoints/48 organizations rises. Check for the indicators Microsoft and the researchers published: unexpected ScreenConnect clients and MSIs at C:\Windows\Temp\sc.msi, RelayTrap files, the RelayTrapLiteAlert Run value, scheduled tasks, and Run Command and Shared Toolbox activity. Review RMM+ versions 1.0.31/1.0.32 and extension configuration. Confirm technician clients are at or above version 26.6.5, and treat web-session exposure on technician hosts as separate from client cleanup. Source recommendations also include not opening interactive sessions with suspect endpoints and using separate security tools for investigation.
- Medium-Term Posture (1–12 months): Monitor for follow-on use of the access (hands-on-keyboard activity, ransomware staging, data theft), for reuse of SCKit or RelayTrap by other actors, and for similar abuse of other remote-support tools. Track vendor disclosures on extension authentication and any change to the ScreenConnect extension review process.
- Scenario Outlook: Most likely: the campaign remains an access-persistence event with limited confirmed technician compromise; trigger: continued reports of failed takeover and no new bypass. Best: cleanup completes and the actor's infrastructure is disrupted; trigger: takedown of attacker-controlled ScreenConnect servers and no second-stage activity. Worst: a bypass or an alternative technician-session route is found, or persisting clients are used for ransomware or data theft; trigger: reports of hijacked console sessions, Toolbox-distributed payloads reaching additional organizations, or extortion claims tied to affected MSPs.
7. Key Individuals and Entities
| Name | Role / Affiliation | Relevance to Assessment |
|---|---|---|
| Unidentified threat actor(s) | Operators of SCKit; attributed to no named group or state | Exploited cloud-hosted ScreenConnect instances beginning November 11, 2026. Identity and objective are not reported. |
| ConnectWise | Vendor of ScreenConnect (formerly ConnectWise Control) | Reported the campaign, issued the advisory on RMM+ 1.0.31/1.0.32, disabled five third-party extensions across about 400 partner environments, and says the CVE-2026-84869 fix blocks the endpoint-to-technician spread. |
| Huntress | Security vendor | Source of the 288-endpoint/48-organization figure and the RMM+ empty-key analysis; says RelayTrap appears LLM-generated and saw console takeover fail in every recovered log. |
| Blackpoint Cyber (Adversary Pursuit Group; researchers Sam Decker and Nevan Beal) | Security vendor | Recovered the loader from attacker infrastructure and decrypted later stages. Reproduced the exploit against a current client and found it blocked. |
| Microsoft Threat Intelligence | Vendor threat research team | Confirmed ConnectWise's reporting, detailed the attack chain (including persistence and key derivation), and published the IOCs and Defender XDR hunting queries. |
| MSP technicians and MSP customers (48 affected organizations) | Victim population | Customer endpoints were compromised; technician workstations are the toolkit's design target for onward spread. |
| RMM+ extension (versions 1.0.31, 1.0.32) | Third-party ScreenConnect extension | Contains the authentication flaw that accepts an empty access key. The developer is not named in available reporting. |
8. Thematic Tags
Cybersecurity, ScreenConnect, managed service providers, supply-chain compromise, remote access abuse, worm malware, third-party extensions, authentication flaw
Explore more: Cybersecurity Briefs · Daily Summary · Support us
✓ YES Dissemination
✓ Cleared Analyst review
| Source | SCI | Role |
|---|---|---|
| bleepingcomputer | 4 | SOURCE_DOCUMENT |
| Microsoft Security Blog | 4 | SOURCE_DOCUMENT |
| Microsoft.com | 4 | SOURCE_DOCUMENT |