# Transparent Tribe (APT36) AI-Assisted Vibeware Campaign — 14+ Malware Families Across 6+ C2 Channels

> Pakistan-nexus threat group APT36 (Transparent Tribe) has launched a sophisticated AI-assisted campaign deploying over 14 malware families written in uncommon languages (Nim, Zig, Crystal, Rust, Go) across 7 distinct C2 channels including Discord, Slack, Google Sheets, Supabase, and Firebase. Dubbed 'Vibeware' by Bitdefender researchers, the campaign targets Indian government, defense, diplomatic missions, and Afghan government entities using a Distributed Denial of Detection (DDoD) strategy that floods environments with disposable polyglot binaries.

- **Published:** 2026-03-07T04:42:47Z
- **Last reviewed:** 2026-03-07T04:42:47Z
- **Canonical:** https://intel.threadlinqs.com/threat/TL-2026-0189
- **Card image:** https://intel.threadlinqs.com/og/threat/threat-intel-transparent-tribe-vibeware-campaign-tl-2026-0189.webp?v=bd24442f (Threadlinqs threat intel: APT36 Transparent Tribe 7 C2 Channels: Detection & IOCs (TL-2026-0189) — High severity, 31 IOCs, 9 detection rules.)
- **ID:** TL-2026-0189
- **Severity:** HIGH
- **Category:** APT
- **Status:** MONITORING
- **Actor:** APT36 (Pakistan)
- **Detections:** 9 (1 shown free below) · **IOCs:** 31 (3 shown free below)

## Description

Transparent Tribe (APT36), a Pakistan-aligned advanced persistent threat group, has adopted an AI-driven malware industrialization model that Bitdefender researchers term 'Vibeware.' Rather than pursuing technical sophistication, the group leverages AI coding assistants to mass-produce malware implants in uncommon programming languages — Nim, Zig, Crystal, Rust, and Go — to evade signature-based detection engines.

The campaign deploys 14+ distinct malware families organized into functional categories:

**Shellcode Loaders:** Warcode (Crystal-based Havoc C2 loader), NimShellcodeLoader (Nim-based Cobalt Strike wrapper using AES-CBC with SHA256 of password 'Pun7sh3r@123'), and ZigLoader (Zig shellcode loader with rolling XOR scheme, base key 0xAA).

**Infostealers:** SheetCreep (C# backdoor using Google Sheets for C2 with Base64+DES-ECB encryption), MailCreep (Go-based infostealer leveraging Microsoft Graph API), LuminousStealer (Rust-based file harvester exfiltrating .txt/.docx/.pdf/.png/.jpg/.xlsx/.pptx/.zip/.rar/.doc/.xls via Firebase and Google Drive), and LuminousCookies (Rust-based Chromium App-Bound Encryption bypass for credential theft).

**Backdoors:** SupaServ (Rust backdoor with Supabase primary and Firebase fallback C2), CrystalShell (Crystal cross-platform backdoor for Windows/Linux/macOS using Discord and Slack C2), ZigShell (Zig-based Slack C2 agent), CrystalFile (Crystal command interpreter monitoring local directories), and Gate Sentinel Beacon (modified open-source C2 framework).

**Support Tools:** CreepDropper (.NET dropper delivering SheetCreep and MailCreep), BackupSpy (Rust filesystem and external media monitoring utility).

The attack chain initiates through spear-phishing emails containing ZIP/ISO archives with LNK shortcuts masquerading as legitimate documents (e.g., resumes, government correspondence). Opening the LNK triggers PowerShell fileless execution via mshta.exe, leading to primary backdoor deployment. The group uses LinkedIn to identify and profile high-value targets in Indian military and diplomatic domains.

C2 infrastructure leverages 7 distinct legitimate cloud services: Discord (hardcoded channel IDs with token progression from hardcoded to Firebase to Supabase), Slack (primary for ZigShell), Google Sheets (SheetCreep hub with columns: unenc_requests, unenc_outputs, unenc_heartbeats, unenc_systems), Supabase (PostgreSQL-based), Firebase Realtime Database, Google Drive (OAuth-authenticated exfiltration), and Microsoft Graph API (MailCreep exfiltration). This multi-channel approach provides operational redundancy through T1008 Fallback Channels.

The 'malware-a-day' production cadence generates daily variants, with compilation timestamps spanning October 2025 through February 2026. Build environment forensics reveal a Rust development path C:\Users\kumar\.cargo\ and the recurring username 'Nightmare' across attacker systems. Unicode emojis embedded in Rust binaries and verbose AI-style comments in source code confirm generative AI tool usage.

Bitdefender characterizes this as a 'Distributed Denial of Detection (DDoD)' strategy — flooding defensive engines with volume and linguistic diversity rather than bypassing them through advanced evasion. While the resulting malware is often unstable and riddled with logical errors, the sheer volume challenges SOC teams and signature-based detection engines.

Payload hosting infrastructure uses domains impersonating Indian High Commission services: hciaccounts.in, hcisupport.in, hcidelhi.in, hcidoc.in, and coadelhi.in. Additional C2 infrastructure includes slackin.online (abusing Azure Front Door) and multiple Firebase Realtime Database instances.

---

**Revalidated on 2026-03-12**

Revalidation confirms this threat remains highly active with significant escalation across targeting scope, malware arsenal, and operational tempo. Since initial publication, several major developments warrant updating the intelligence record.

**Expanded Targeting Beyond Government/Defense:** Acronis TRU documented a strategic pivot in January 2026 where APT36 expanded beyond traditional government/defense targets to India's startup ecosystem, specifically OSINT and cybersecurity firms, using ISO-delivered Crimson RAT with personalized startup-founder-themed lure material. Indian embassies in multiple foreign countries and Afghan government entities have also been confirmed as active targets via LinkedIn-based reconnaissance.

**Operation Sindoor (May 2025):** Seqrite Labs documented a coordinated cyber siege where APT36 collaborated with 35+ hacktivist outfits, producing 650+ cyber incidents between May 7-10, 2025. The operation targeted Indian critical infrastructure including telecom providers (Jio, BSNL), healthcare institutions (AIIMS, Apollo Hospitals), defense entities (MoD, Army, Navy, DRDO), and state education portals. This represents APT36's most significant known coordination with non-state hacktivist actors and signals convergence of espionage and ideological warfare.

**Malware Arsenal Growth:** The observable malware family count has grown from the initial 14 to 18+, with newly documented families including GETA RAT (.NET-based RAT linked to SideCopy cluster), Ares RAT (Python-based, Go-delivered Linux RAT), DeskRAT (WebSocket-based C2 with real-time host telemetry), and Poseidon (Linux Mythic framework agent with keylogging, screen capture, and file exfiltration). The Aryaka C2 tradecraft report (February 2026) provided the most comprehensive network-level analysis to date, documenting fixed-size encrypted command packets, regular beacon intervals, and long-lived outbound connections across non-standard ports.

**Pahalgam Terror Attack Exploitation:** Seqrite Labs revealed that APT36 pre-compiled three RAT payloads on April 21, 2025 — one day before the Pahalgam terror attack — then weaponized the tragedy as phishing lure material with fake domains impersonating Jammu & Kashmir Police (jkpolice.gov.in.kashmirattack.exposed) and Indian Air Force, delivering Crimson RAT via PPAM macros.

**Advanced LNK Tradecraft:** CYFIRMA and PolySwarm documented evolution of APT36's LNK-based delivery mechanism: oversized 2MB+ shortcut files embedding complete PDF documents, fileless execution via mshta.exe, layered in-memory decryption, and antivirus-aware persistence mechanisms that adapt behavior based on the security product environment detected on the target host.

**Linux Targeting Expansion:** APT36 is increasingly targeting Linux environments due to Indian government adoption of Debian-based BOSS OS and the introduction of Maya OS, deploying Poseidon and Ares RAT via Go-based downloaders and shell scripts.

**Government Response:** CERT-In and Chandigarh Police issued renewed warnings about APT36 activity. DSCI published a formal threat advisory in October 2025. An academic paper on arXiv documented the Operation Sindoor malware campaign analysis and proposed detection frameworks.

## MITRE ATT&CK

- T1589 Gather Victim Identity Information
- T1583 Acquire Infrastructure
- T1585 Establish Accounts
- T1587 Develop Capabilities
- T1588 Obtain Capabilities
- T1608 Stage Capabilities
- T1566 Phishing
- T1059 Command and Scripting Interpreter
- T1204 User Execution
- T1129 Shared Modules
- T1106 Native API
- T1053 Scheduled Task/Job
- T1547 Boot or Logon Autostart Execution
- T1140 Deobfuscate/Decode Files or Information
- T1036 Masquerading
- T1027 Obfuscated Files or Information
- T1620 Reflective Code Loading
- T1564 Hide Artifacts
- T1218 System Binary Proxy Execution
- T1056 Input Capture
- T1187 Forced Authentication
- T1033 System Owner/User Discovery
- T1087 Account Discovery
- T1005 Data from Local System
- T1530 Data from Cloud Storage
- T1115 Clipboard Data
- T1560 Archive Collected Data
- T1071 Application Layer Protocol
- T1102 Web Service
- T1573 Encrypted Channel
- T1132 Data Encoding
- T1008 Fallback Channels
- T1105 Ingress Tool Transfer
- T1041 Exfiltration Over C2 Channel
- T1048 Exfiltration Over Alternative Protocol

## Free detection (Splunk SPL, Microsoft KQL, Sigma)

**APT36 Vibeware — mshta.exe Spawning PowerShell via LNK Execution** (`TL-2026-0189-SPL-001`)

Detects mshta.exe spawning PowerShell or cmd.exe with hidden window and execution policy bypass parameters, matching APT36 Vibeware initial access chain where malicious LNK files in ZIP/ISO archives trigger mshta.exe to execute remote HTA payloads from payload hosting domains (hciaccounts.in, hcisupport.in), which then spawn hidden PowerShell for payload delivery and system reconnaissance.

- **Languages:** Splunk SPL, Microsoft KQL, Sigma
- **ATT&CK:** T1204.002, T1218.011, T1059.001
- **Data sources:** sysmon, windows_security, edr
- **Severity:** high
- **Confidence:** high

### Splunk SPL

```spl
index=sysmon sourcetype="XmlWinEventLog:Microsoft-Windows-Sysmon/Operational" EventCode=1 ParentImage="*\\mshta.exe" (Image="*\\powershell.exe" OR Image="*\\cmd.exe") (CommandLine="*-WindowStyle Hidden*" OR CommandLine="*-ExecutionPolicy Bypass*" OR CommandLine="*-enc*" OR CommandLine="*-w hidden*") | eval risk_score=if(match(CommandLine,"(?i)-enc"),90,80) | table _time ComputerName User ParentImage ParentCommandLine Image CommandLine risk_score | sort -_time
```

### Microsoft KQL

```kql
DeviceProcessEvents | where Timestamp > ago(24h) | where InitiatingProcessFileName =~ "mshta.exe" | where FileName in~ ("powershell.exe", "cmd.exe") | where ProcessCommandLine has_any ("-WindowStyle Hidden", "-ExecutionPolicy Bypass", "-enc", "-w hidden") | project Timestamp, DeviceName, AccountName, InitiatingProcessFileName, InitiatingProcessCommandLine, FileName, ProcessCommandLine | sort by Timestamp desc
```

### Sigma

```yaml
title: APT36 Vibeware — mshta.exe Spawning PowerShell via LNK Execution
id: d1f6a4e5-7b8c-9d0e-1f2a-3b4c5d6e7f80
status: experimental
description: Detects mshta.exe spawning PowerShell or cmd.exe with hidden window parameters, matching APT36 Vibeware initial access chain where malicious LNK files trigger mshta.exe to execute remote HTA payloads.
references:
    - https://www.bitdefender.com/en-us/blog/businessinsights/apt36-nightmare-vibeware
    - https://thehackernews.com/2026/03/transparent-tribe-uses-ai-to-mass.html
author: Threadlinqs Team
date: 2026/03/07
tags:
    - attack.execution
    - attack.t1059.001
    - attack.defense_evasion
    - attack.t1218.011
    - attack.initial_access
    - attack.t1204.002
logsource:
    product: windows
    category: process_creation
detection:
    selection_parent:
        ParentImage|endswith: '\mshta.exe'
    selection_child:
        Image|endswith:
            - '\powershell.exe'
            - '\cmd.exe'
    selection_cmdline:
        CommandLine|contains:
            - '-WindowStyle Hidden'
            - '-ExecutionPolicy Bypass'
            - '-enc'
            - '-w hidden'
    condition: selection_parent and selection_child and selection_cmdline
falsepositives:
    - Legitimate IT administration tools using mshta.exe for scripted deployments
level: high
```

**Known false positives:**

- Legitimate IT administration tools using mshta.exe for scripted deployments
- Application packaging tools using mshta.exe as a wrapper

**Tuning notes:** Exclude known IT admin tools by adding ParentCommandLine filters for legitimate mshta.exe usage. Adjust risk_score thresholds based on environment baseline. Consider adding allowlists for signed parent processes.

### 8 more detection rules (Splunk SPL, Microsoft KQL, Sigma), unlock with TL Creds

- APT36 Vibeware — LNK File Triggering Script Interpreter Execution (locked: unlock with TL Creds)
- APT36 Vibeware — Unsigned Polyglot Binary Execution from User-Writable Directories (locked: unlock with TL Creds)
- APT36 Vibeware — Sensitive File Staging and Exfiltration via Cloud APIs (locked: unlock with TL Creds)
- APT36 Vibeware — Connections to Payload Hosting Domains (locked: unlock with TL Creds)
- APT36 Vibeware — Campaign-Specific Persistence Artifacts (locked: unlock with TL Creds)
- APT36 Vibeware — Known Firebase and Slack C2 Infrastructure Connections (locked: unlock with TL Creds)
- APT36 Vibeware — Scheduled Task Masquerading as Edge Update with Registry Persistence (locked: unlock with TL Creds)
- APT36 Vibeware — Multi-Channel Cloud C2 Beaconing from Non-Browser Process (locked: unlock with TL Creds)

Sign up free to unlock the detection rules: https://threadlinqs.com/?modal=signup

## Sample indicators (3 of 31)

Defanged, TLP:CLEAR.

- IP address: `23.152.0[.]81`
- Domain: `slackin[.]online`
- File hash: `b56062033df06738b66c38b3fa2f82a7e8c558336a4790c83c7faad595172267`

Full set of 31 indicators unlocks with TL Creds: https://threadlinqs.com/tl-creds

## Sources

- [APT36: A Nightmare of Vibeware](https://www.bitdefender.com/en-us/blog/businessinsights/apt36-nightmare-vibeware)
- [Transparent Tribe Uses AI to Mass-Produce Malware Implants in Campaign Targeting India](https://thehackernews.com/2026/03/transparent-tribe-uses-ai-to-mass.html)
- [APT Attacks Target Indian Government Using SHEETCREEP, FIREPOWER, and MAILCREEP](https://www.zscaler.com/blogs/security-research/apt-attacks-target-indian-government-using-sheetcreep-firepower-and)
- [Nation-State Actor Embraces AI Malware Assembly Line](https://www.darkreading.com/cyberattacks-data-breaches/nation-state-actor-ai-malware-assembly-line)
- [Pakistan-Linked APT36 Floods Indian Govt Networks With AI-Made Vibeware](https://hackread.com/pakistan-apt36-indian-govt-networks-ai-vibeware/)
- [Nation-State Hackers Play the Vibes](https://www.bankinfosecurity.com/nation-state-hackers-play-vibes-a-30920)
- [Transparent Tribe Launches New RAT Attacks Against Indian Government and Academia](https://thehackernews.com/2026/01/transparent-tribe-launches-new-rat.html)
- [APT36 and SideCopy Launch Cross-Platform RAT Campaigns Against Indian Entities](https://thehackernews.com/2026/02/apt36-and-sidecopy-launch-cross.html)
- [Transparent Tribe APT Infrastructure Mapping](https://www.team-cymru.com/post/transparent-tribe-apt-infrastructure-mapping)
- [APT36 Phishing Campaign Targets Indian Defense Using Credential-Stealing Malware](https://www.cyfirma.com/research/apt36-phishing-campaign-targets-indian-defense-using-credential-stealing-malware/)

## Full data

The other 8 detection rules (Splunk SPL / Microsoft KQL / Sigma) unlock with TL Creds; a free account includes daily creds: https://threadlinqs.com/?modal=signup
Indicator values unlock with TL Creds, at a lower price on Red and Purple; AI agents can query rules and IOCs through the Threadlinqs MCP server (Purple): https://intel.threadlinqs.com/mcp

Canonical: https://intel.threadlinqs.com/threat/TL-2026-0189
