CVE-2026-20131 — Interlock Ransomware Exploiting Cisco Secure Firewall Management Center RCE

Insecure deserialization of a user-supplied Java byte stream on the Cisco Secure FMC web-management interface lets an unauthenticated attacker execute arbitrary Java code as root via crafted HTTP requests. CVSS 10.0. Exploited as a zero-day by the Interlock ransomware group since 2026-01-26 — 36 days before Cisco's 2026-03-04 disclosure. Defensive hunt & harden pack — no exploit code.
Threat
CVE-2026-20131 · Interlock · Cisco Secure FMC
Severity
EXPLOITED ITW · CVSS 10.0 · ROOT RCE
Type
Insecure deserialization (pre-auth RCE) → ransomware
Patch
Cisco FMC fixed release (per Cisco SA, 2026-03-04)
Actor
Interlock ransomware group (double extortion)
Version
v0.1 · 2026-06-13
Author
HuntPack
01

Executive Summary

CVE-2026-20131 is a CVSS 10.0 remote-code-execution flaw in Cisco Secure Firewall Management Center (FMC). The web-based management interface insecurely deserializes a user-supplied Java byte stream, allowing an unauthenticated attacker to execute arbitrary Java code as root via crafted HTTP requests. FMC is the centralized brain for an organization's Cisco firewall estate, so root on it is a strategic foothold.

The Interlock ransomware group exploited this as a zero-day from 2026-01-26 — roughly 36 days before Cisco's official disclosure on 2026-03-04 — per Amazon threat-intelligence and eSentire reporting. Post-exploitation, the actor runs multi-stage intrusions: reconnaissance, payload deployment and persistence, abusing legitimate tooling including ScreenConnect RMM, the Volatility memory-forensics suite (credential / memory access), and the offensive AD-CS tool Certify. Targeted sectors include education, engineering / architecture / construction, manufacturing, healthcare and government.

Edge-device priority: Upgrade FMC to the Cisco fixed release (S10) — that closes the RCE. The FMC appliance is not a Falcon-sensored host, so detection splits: (1) hunt the FMC web/admin logs if ingested into LogScale / NG-SIEM (Q1–Q2 — anomalous deserialization-style POSTs, unexpected root/admin activity), and (2) hunt internal endpoints for the Interlock post-exploit kill-chain (Q3–Q6 — ScreenConnect/RMM, Certify / AD-CS abuse, recovery inhibition, encryption). Highest-fidelity endpoint signals: Q4 (Certify) and Q5 (recovery inhibition).
02

Source Review & Web Hunter Notes

TierSourceKey FindingCarry
1 · Vendor (PSIRT)Cisco Security Advisory — CVE-2026-20131Insecure Java deserialization on FMC web UI; unauth RCE as root; fixed release 2026-03-04yes
1 · Vendor researchAmazon (AWS) Threat IntelligenceIdentified the active Interlock campaign targeting enterprise firewalls via this flawyes
1 · Vendor researcheSentire — CVE-2026-20131 exploited by InterlockCrafted HTTP requests to FMC web mgmt; multi-stage intrusion; ScreenConnect / Volatility / Certify abuseyes
1 · News of recordThe Hacker News — Interlock exploits Cisco FMC zero-day for rootZero-day since 2026-01-26, 36 days before disclosure; root access; sectors targetedyes
2 · Vendor researchRecorded Future / Hive Pro / Help Net SecurityMarch 2026 CVE landscape; FMC zero-day exploited weeks before patch; CVSS 10.0partial
2 · Vendor researchSecurity Affairs / SecPodEarly-access exploitation timeline; deserialization root-cause detailpartial

Decisions: lead with the upgrade (the Cisco fixed release closes the RCE). The appliance is off-sensor, so durable detections are (a) FMC web/admin log hunts where ingested and (b) endpoint hunts for Interlock's hands-on-keyboard tooling, which is the more reliable signal in most Falcon estates. Key on tooling/behavior (ScreenConnect, Certify, recovery inhibition, encryption) since exploit-stage atomic IOCs are perishable and not fully public — pull live IOCs from the eSentire / Amazon / Cisco advisories.

03

Hunt Brief & Attack Chain

Working hypothesis: An attacker sends crafted HTTP requests to an internet-reachable Cisco Secure FMC web interface, triggers insecure Java deserialization, and gains root code execution. From that foothold Interlock pivots into the managed network, deploys ScreenConnect for hands-on access, uses Certify / Volatility for credential and AD-CS abuse, inhibits recovery, and deploys ransomware for double extortion.

StepBehaviorTelemetryHunt Angle
1 · Exploit FMCCrafted HTTP request → insecure Java deserialization → root code exec (CVE-2026-20131)FMC web/admin logsQ1, Q2; Native N2
2 · Establish accessDeploy ScreenConnect / RMM for interactive control on internal hostsProcessRollup2, NetworkConnectIP4Q3
3 · Credential / AD-CS abuseCertify (AD-CS), Volatility (memory) for credential and certificate theftProcessRollup2Q4
4 · Recon + lateralAD/host enumeration, lateral movement to high-value systemsProcessRollup2, UserLogonBaseline; N4
5 · Inhibit recoveryShadow-copy / backup / recovery deletion before encryptionProcessRollup2Q5
6 · ImpactHigh-volume file encryption, ransom note, double-extortion leakFileCreateInfoQ6; §11

Affected surface & telemetry

SurfaceRequired TelemetryPriorityGap Risk
Internet-reachable Cisco Secure FMC web UIFMC version inventory + web/admin access logs ingestedCriticalHigh — appliance off-sensor; depends on log forwarding
RMM / remote-access toolingProcess exec + network (ProcessRollup2, NetworkConnectIP4)HighLow — standard Falcon sensor
AD-CS / credential abuseProcess exec + command line (ProcessRollup2)HighMedium — renamed tooling evades name match
Recovery inhibition / encryptionProcess + file ops (ProcessRollup2, FileCreateInfo)HighLow — standard Falcon sensor
04

Consolidated IOC Table

The Cisco fixed release is the definitive control; the strongest detections are behavioral (hands-on tooling). Exploit-stage atomic IOCs (FMC payload hashes, exploit infra) are perishable and not fully public — pull live IOCs from the eSentire / Amazon / Cisco advisories. Copy-ready blocks in §9.
TypeValueConfActionContext
VersionCisco Secure FMC below the 2026-03-04 fixed releasehighdetectVulnerable to CVE-2026-20131 — upgrade closes it
BehaviorAnomalous deserialization-style POST to FMC web mgmt + root process spawnmediumhuntExploitation signature (log-dependent)
ToolScreenConnect / unsanctioned RMM on internal hostshighdetectInterlock interactive access
ToolCertify.exe / AD-CS abuse; Volatility on production hostshighdetectCredential / certificate theft
Behaviorvssadmin/wbadmin/bcdedit recovery deletion → mass file writehighdetectInterlock pre-encryption kill-chain
05

ATT&CK Mapping

TacticTechniqueObserved BehaviorQuery / Control
Initial AccessT1190 — Exploit Public-Facing ApplicationInsecure deserialization RCE on the FMC web UIQ1, Q2; S10 upgrade
ExecutionT1059 — Command and Scripting InterpreterArbitrary Java code as root post-deserializationQ2
Persistence / C2T1219 — Remote Access SoftwareScreenConnect / RMM for hands-on accessQ3
Credential AccessT1649 — Steal or Forge Authentication CertificatesCertify abuse against AD-CSQ4
Credential AccessT1003 — OS Credential DumpingVolatility for memory/credential extractionQ4
ImpactT1490 / T1486 — Inhibit Recovery / Data EncryptedShadow deletion then encryptionQ5, Q6; §11
06

Native Audit-Log Hunts

HuntSourceLogicResponse
N1 · FMC version / patch inventoryCisco FMC / asset inventoryFMC instances below the 2026-03-04 fixed release, especially internet-reachable web UIsUpgrade to the fixed release — the fix
N2 · FMC web/admin log reviewFMC system & audit logsAnomalous POSTs to the web mgmt interface, unexpected admin/root actions since 2026-01-26 (zero-day window)Per Cisco/eSentire guidance; pivot to endpoint hunts
N3 · Sanctioned-RMM allow-list reviewSoftware inventory / EDRScreenConnect or other RMM present on hosts where it is not sanctionedInvestigate; remove; scope access
N4 · AD-CS abuse reviewAD-CS / Windows Security logs (4886/4887)Unusual certificate enrollment / template abuse following the FMC windowCorrelate with Q4; revoke abused certs
07

CrowdStrike LogScale CQL Hunt Queries

Pick your tenant's cloud first — every "Open in Falcon" button below uses this selection.
Q1–Q2 run against ingested Cisco FMC web/admin logs — field/parser names vary by integration, so treat those as ⚠ validate in your tenant. Q3–Q6 are Falcon-sensor endpoint hunts (fields validated against the Falcon event reference) covering the Interlock post-exploit kill-chain.
Q1 · Anomalous POST to the FMC web management interface (deserialization)
CONF MEDFP MEDCOST LOW

Looks for: unauthenticated POST requests to FMC web-management endpoints carrying serialized-object-style payloads (Java rO0 base64 marker, large bodies, odd content types). ⚠ field names parser-dependent

// HUNT: Anomalous deserialization-style POST to the Cisco FMC web UI (CVE-2026-20131)
// MITRE: T1190 | CONF: medium  FP: medium  COST: low
// REQUIRES: Cisco FMC web/admin/HTTP logs ingested into LogScale/NG-SIEM
// FALSE POSITIVES: legitimate large admin POSTs / API automation
// TUNING: scope to the FMC mgmt VIP; baseline known API clients; alert on unauth + rO0 marker
vendor=/cisco/i product=/Firewall Management Center|FMC/i
| http_method="POST"
| url=/\/(api|admin|ui|webui)\//i
| (request_body=/rO0[AB]/ OR content_type=/application\/x-java-serialized-object/i OR body_len > 50000)
| auth_status=/fail|none|anonymous/i
| table([@timestamp, src_ip, http_method, url, content_type, body_len, user_agent], limit=200)
| sort(@timestamp, order=desc)
Q2 · Unexpected root/shell activity on the FMC appliance
CONF MEDFP LOW-MEDCOST LOW

Looks for: shell / network-tool execution as root on the FMC host correlated to the web service — the post-deserialization payload. ⚠ field names parser-dependent

// HUNT: Root shell / outbound tooling on the FMC appliance after web-UI activity
// MITRE: T1190, T1059 | CONF: medium  FP: low-med  COST: low
// REQUIRES: FMC OS/syslog (auth, process, or audit) ingested
// FALSE POSITIVES: sanctioned maintenance windows / Cisco TAC sessions
// TUNING: allow-list scheduled maintenance; alert on shells parented by the web/Java service
vendor=/cisco/i host=/fmc/i
| user=/root/i
| process=/(bash|sh|curl|wget|nc|ncat|python|perl)$/i
| parent_process=/(java|tomcat|httpd|nginx|mojo)/i
| table([@timestamp, host, user, process, parent_process, cmdline], limit=200)
| sort(@timestamp, order=desc)
Q3 · Unsanctioned RMM / remote-access tooling (ScreenConnect)
CONF HIGHFP MEDCOST LOW

Looks for: ScreenConnect / common RMM binaries executing on internal hosts. FP: sanctioned IT RMM — allow-list your approved tool and management hosts.

// HUNT: Unsanctioned RMM / remote-access software (Interlock hands-on access)
// MITRE: T1219 | CONF: high  FP: medium  COST: low
// REQUIRES: ProcessRollup2
// FALSE POSITIVES: the org's sanctioned RMM tool / IT admin hosts
// TUNING: allow-list the approved RMM product + its deployment hosts; alert on the rest
#event_simpleName=/ProcessRollup2|SyntheticProcessRollup2/
| FileName=/^(ScreenConnect|Connectwisechat|AnyDesk|atera|splashtop|ScreenConnect\.ClientService|ScreenConnect\.WindowsClient)\.exe$/i
| table([@timestamp, ComputerName, UserName, FileName, CommandLine, ParentBaseFileName, SHA256HashData], limit=200)
| sort(@timestamp, order=desc)
Q4 · AD-CS / credential-theft tooling (Certify, Volatility)
CONF HIGHFP LOWCOST LOW

Looks for: offensive AD-CS tooling (Certify) and memory-forensics (Volatility) on production hosts, plus their tell-tale command-line verbs. FP: sanctioned red-team / IR — allow-list.

// HUNT: Certify / Volatility credential & certificate-theft tooling on production hosts
// MITRE: T1649, T1003 | CONF: high  FP: low  COST: low
// REQUIRES: ProcessRollup2
// FALSE POSITIVES: sanctioned red-team / IR engagements — allow-list operator hosts/AIDs
#event_simpleName=/ProcessRollup2|SyntheticProcessRollup2/
| (FileName=/^(Certify|vol|volatility|volatility3)\.exe$/i
   OR CommandLine=/certify\s+(find|request|enroll)|--vulnerable|windows\.pslist|windows\.lsadump/i)
| table([@timestamp, ComputerName, UserName, FileName, CommandLine, ParentBaseFileName, SHA256HashData], limit=200)
| sort(@timestamp, order=desc)
Q5 · Recovery inhibition — shadow copy / backup deletion
CONF HIGHFP LOWCOST LOW

Looks for: shadow-copy, backup-catalog and boot-recovery deletion — the Interlock pre-encryption step. Accomplishes: high-fidelity ransomware staging detection independent of the FMC entry path.

// HUNT: Volume shadow / backup / recovery deletion ahead of encryption (Interlock kill-chain)
// MITRE: T1490 | CONF: high  FP: low  COST: low
// REQUIRES: ProcessRollup2 command-line telemetry
// FALSE POSITIVES: sanctioned backup maintenance — allow-list backup service accounts
#event_simpleName=/ProcessRollup2|SyntheticProcessRollup2/
| FileName=/^(vssadmin|wmic|bcdedit|wbadmin|powershell|cmd)\.exe$/i
| CommandLine=/(delete\s+shadows|shadowcopy\s+delete|resize\s+shadowstorage|deletecatalog|recoveryenabled\s+no|bootstatuspolicy\s+ignoreallfailures)/i
| table([@timestamp, ComputerName, UserName, FileName, CommandLine, ParentBaseFileName], limit=200)
| sort(@timestamp, order=desc)
Q6 · High-volume file modification by a single process (encryption)
CONF MEDFP MEDCOST MED

Looks for: one process touching an abnormally high number of distinct files in a short window — the encryption burst. FP: backup/indexing/AV — allow-list. Pair with Q5 on the same host for confidence.

// HUNT: Single process performing high-volume file modification (ransomware encryption burst)
// MITRE: T1486 | CONF: medium  FP: medium  COST: medium
// REQUIRES: FileCreateInfo / file-write telemetry with writing-process context
// FALSE POSITIVES: backup agents, search indexers, AV — allow-list by process
// TUNING: tune the threshold to your environment; correlate with Q5 hits on same aid
#event_simpleName=/PeFileWritten/
| FileName!=/^(Backup|SearchIndexer|CSFalconService|MsMpEng)\.exe$/i
| groupBy([aid, ContextProcessId, ContextBaseFileName], function=([
    count(TargetFileName, distinct=true, as=filesTouched),
    min(@timestamp, as=t0), max(@timestamp, as=t1)
  ]))
| filesTouched >= 200
| sort(filesTouched, order=desc)
08

CrowdStrike Custom IOA Recommendations

IOA NameField PatternsBenign ExclusionsDeployment Path
Interlock — AD-CS theft toolingImage Certify.exe or CmdLine certify find/request / --vulnerableSanctioned red-team operator hosts (excluded)Endpoint Security → Custom IOA → Process Creation (Detect+Block)
Interlock — Recovery inhibitionChild vssadmin/wbadmin/bcdedit.exe; CmdLine delete shadows / recoveryenabled noSanctioned backup service accounts (excluded)Custom IOA → Process Creation (Detect+Block)
Q4 and Q5 are the strongest promotions (almost no benign baseline). Q3 (RMM) stays Detect-only because sanctioned RMM exists — promote per-host once your approved tool is allow-listed. Q1–Q2 run on FMC logs (off-sensor) → NG-SIEM scheduled searches, not Custom IOAs. Q6 stays hunt/investigate; correlate before alerting.
09

Machine-Readable IOC Appendix

Grouped IOC Quick-Copy

One-click blocks for version audit, behavioral detection and tooling allow-listing. The Cisco fixed release is the control; behavior + tooling are the durable detections. No complete public atomic IOC set was released — pull live exploit IOCs from the eSentire / Amazon / Cisco advisories.

Behavioral Signaturesdetection logic
Anonymous deserialization-style POST to the FMC web mgmt interface (rO0 marker / java-serialized)
Root shell or curl/wget/nc parented by the FMC web/Java service
ScreenConnect / unsanctioned RMM executing on internal hosts
Certify (AD-CS) or Volatility (memory) tooling on production hosts
vssadmin/wbadmin/bcdedit recovery-deletion commands (Interlock pre-encryption)
One process modifying 200+ distinct files in under a minute (encryption burst)
FMC Version / Patch Auditthe fix
# Cisco Secure FMC — confirm patched against CVE-2026-20131
# 1) Identify FMC software version (Help > About / System > Configuration)
# 2) Upgrade to the Cisco fixed release referenced in the 2026-03-04 advisory
# 3) Restrict the FMC web-management interface to a management network ONLY
#    - never expose the FMC web UI to the internet
# 4) Review FMC audit logs back to 2026-01-26 (zero-day window) for prior compromise
# Reference: Cisco Security Advisory CVE-2026-20131 (CVSS 10.0)
Tooling Allow-List / BlockRMM & offensive tools
# Interlock hands-on-keyboard tooling to allow-list (sanctioned) or block (unexpected)
ScreenConnect / ConnectWise Control   # RMM — keep ONLY your approved tool
AnyDesk / Atera / Splashtop           # other RMM seen in extortion intrusions
Certify.exe                           # offensive AD-CS abuse — block on production
volatility / vol.py                   # memory forensics — block outside IR hosts
# Enforce: application control / RMM allow-listing; alert on any RMM not on the list
Related ReferencesCVE / advisory
CVE-2026-20131   # Cisco Secure FMC insecure-deserialization RCE (this pack) — CVSS 10.0
Interlock        # ransomware group; double extortion; exploited the FMC zero-day
# Timeline: zero-day exploitation since 2026-01-26; Cisco disclosure 2026-03-04
# Live IOCs: pull from eSentire / Amazon (AWS) Threat Intelligence / Cisco advisories
# Defense-in-depth: isolate FMC mgmt plane; RMM allow-listing; AD-CS hardening
10

Hardening — Tiered & Deployable

This one is patchable — upgrading FMC is the fix. The rest isolates the management plane and removes the Interlock hands-on toolkit's room to operate.

Immediate — Close the RCE (the fix)
  • Upgrade Cisco Secure FMC to the fixed release (per the 2026-03-04 advisory) — M1051. Closes CVE-2026-20131.
  • Remove the FMC web-management interface from internet exposureM1030 / M1035. Restrict to a dedicated management network / jump host.
  • Review FMC audit logs back to 2026-01-26M1047. Confirm no prior zero-day compromise before declaring closed.
  • RMM allow-listing — block all RMM except the sanctioned toolM1038 / M1042. Defeats the ScreenConnect hands-on stage (Q3).
Near term — Constrain the post-exploit toolkit (pilot first)
  • Harden AD-CS against Certify-style abuseM1042 / M1018. Remove ESC-vulnerable templates; enforce manager approval; restrict enrollment agents.
  • Application control to block offensive tooling on productionM1038. Stops Certify / Volatility outside sanctioned IR hosts (Q4).
  • Tamper-protected, offline / immutable backupsM1053. Neutralizes recovery inhibition (Q5).
  • Segment management of network appliancesM1030. A root foothold on FMC can't pivot freely into production.
Strategic — Contain blast radius & monitor drift
  • Out-of-band management plane for all security appliancesM1030. Isolate FMC/FTD admin from user networks.
  • Continuous appliance version & config-drift monitoringM1016. Alert when an appliance falls behind on PSIRT fixes or re-exposes its web UI.
  • Default-deny application control + tiered admin modelM1038 / M1018. Raises cost across the whole Interlock kill-chain.
11

Containment Runbook

PhaseActionsOwnerEvidence
IsolateRestrict FMC mgmt access to a break-glass admin path; network-contain internal hosts running unsanctioned RMM or showing Q4/Q5 hitsNetwork + SOC L2Q3–Q5 rows; FMC access log
TriagePull FMC audit logs (2026-01-26 →) for exploitation; map RMM / Certify / Volatility execution and any recovery-deletion activity on endpointsSOC L2FMC logs; Q3–Q6 rows
EradicateRemove planted RMM and tooling; revoke certificates abused via AD-CS; reset credentials harvested via memory/AD-CS; rebuild FMC if root compromise confirmedIRCert / cred revocation log
RecoverRestore from offline backups if encryption occurred; bring FMC up only on the fixed release behind the mgmt network; rotate privileged credentialsIR + ITRestore log; FMC version readout
HardenRMM allow-listing; AD-CS template hardening; promote Q4/Q5 IOAs; enable appliance config-drift monitoring (§10)Detection EngPolicy state; IOA enabled
12

Detection Coverage Map & Validation

TechniqueBehaviorCQLIOACoverage
Vulnerable FMC version / exposed web UIGood upgrade + N1/N2 audit
T1190 / T1059FMC deserialization exploitationQ1, Q2Partial needs FMC logs ingested
T1219RMM / ScreenConnect hands-on accessQ3Partial Detect-only; allow-list first
T1649 / T1003Certify / Volatility credential theftQ4IOA-1Good
T1490Recovery inhibitionQ5IOA-2Good
T1486Encryption burstQ6Partial threshold-tuned

Validation gates: (1) confirm Cisco FMC logs are ingested and the parser field names in Q1–Q2 resolve in your tenant; (2) allow-list your sanctioned RMM before enabling Q3 alerting; (3) Q4/Q5 return zero hits in a 30-day benign lookback; (4) tune the Q6 filesTouched threshold against a benign baseline; (5) inventory FMC versions (N1) and confirm the fixed release + that the web UI is off the internet; (6) promote Q4/Q5 to Custom IOA Detect+Block (target FP < 1/week).

13

Hunt Summary Ticket

TITLE:        Hunt — CVE-2026-20131 (Interlock / Cisco Secure FMC Deserialization RCE)
SEVERITY:     Critical (Exploited ITW, CVSS 10.0, unauth root RCE; Interlock ransomware)
SCOPE:        Cisco Secure FMC (web UI) + internal endpoints reached post-exploit
HYPOTHESIS:   Attacker sends crafted HTTP requests to the FMC web UI, triggers insecure Java
              deserialization for root code exec, then deploys ScreenConnect, abuses AD-CS via
              Certify, inhibits recovery and encrypts (Interlock double extortion).
QUERIES RUN:  Q1 FMC deserialization POST | Q2 FMC root shell | Q3 unsanctioned RMM |
              Q4 Certify/Volatility (IOA) | Q5 recovery inhibition (IOA) | Q6 encryption burst
              + Native N1-N4 (FMC version, FMC log review, RMM allow-list, AD-CS abuse)
DO FIRST:     Upgrade FMC + remove web UI from internet; then Q4, Q5 -> escalate hits
FINDINGS:     <pending analyst execution>
GAPS:         Q1-Q2 depend on Cisco FMC logs being ingested; parser field names vary
ACTIONS:      Patch FMC; isolate mgmt plane; RMM allow-listing; AD-CS hardening;
              promote Q4/Q5 to Custom IOA Detect+Block
OWNER:        HuntPack
VERSION:      v0.1 - 2026-06-13
14

Changelog

v0.32026-07-24CQL correctness pass. All table() calls now carry an explicit row limit: the default is 200 and truncation is silent, so a capped result was indistinguishable from a complete one. Atomic-IOC sweeps (filters over 5+ hashes or C2 IPs) use limit=max so a wide infection is never silently under-scoped; behavioural hunts use limit=200, where exceeding the cap indicates the query needs tuning. Where present, event names that do not exist in the Falcon data model were corrected (e.g. ServiceInstalled is a Sysmon concept, not a Falcon event; ElfFileWritten is ELFFileWritten) — such queries could never return a row. No detection logic, fields, or IOCs changed.
v0.22026-06-29CQL syntax review (crowdstrike-logscale-v3): removed invalid FileCreateInfo event name from the mass-file-write hunt.
v0.12026-06-13Initial pack — built to the review-v2 gold layout (fixed left-sidebar scrollspy TOC, collapsible, Falcon cloud selector inside the CQL section, per-card Copy/Open-in-Falcon, Grouped IOC Quick-Copy grid, coverage+validation gates, .ticket). Covers CVE-2026-20131 Interlock / Cisco Secure FMC insecure-deserialization RCE (CVSS 10.0): 6 CQL hunts (2 FMC-log + 4 endpoint), 4 native hunts, 2 IOA candidates. Keyed on version-state + Interlock hands-on tooling (ScreenConnect, Certify, Volatility) + recovery-inhibition / encryption kill-chain. Exploit-stage atomic IOCs perishable / not fully public — pull live from vendor advisories.
15

References

TierSourceUsed For
1Amazon (AWS) Threat Intelligence — Interlock targeting enterprise firewallsCampaign identification, FMC targeting, IOCs
1eSentire — CVE-2026-20131 exploited by InterlockCrafted-HTTP exploitation, ScreenConnect / Volatility / Certify tooling
1The Hacker News — Interlock exploits Cisco FMC zero-day for rootRoot RCE, zero-day timeline (2026-01-26), targeted sectors
2Help Net Security — Cisco FMC flaw exploited weeks before patchDisclosure timeline, exposure, CVSS 10.0
2Hive Pro — CVE-2026-20131 threat advisoryDeserialization root-cause, mechanism
2Recorded Future — March 2026 CVE landscapeContext; Interlock FMC zero-day in the monthly landscape

HuntPack v0.1 · CVE-2026-20131 (Interlock / Cisco Secure FMC RCE) · Generated 2026-06-13 · Defensive use only — no exploit code. The fix (upgrade FMC to the Cisco fixed release + remove the web UI from the internet) closes the RCE; the hunts are detection & defense-in-depth for the Interlock post-exploit kill-chain. Q1–Q2 field names depend on your Cisco FMC log parser — validate in your tenant before alerting.