---
id: CVE-2026-61732
title: Decepticon is an autonomous hacking agent for red teams
summary: >-
  Decepticon is an autonomous hacking agent for red teams. Versions prior to
  1.1.17 wrap web crawl results — the output of agent reconnaissance against
  target services — into LLM messages without neutralizing ChatML special-token
  literals.…
severity: critical
cvss: 10
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H'
cwe:
  - CWE-74
vendor: decepticon-core
product: decepticon-core
affected:
  - decepticon-core <= 1.1.16
  - decepticon <= 1.1.16
  - decepticon-sdk <= 1.1.16
patched:
  - decepticon-core 1.1.17
  - decepticon 1.1.17
  - decepticon-sdk 1.1.17
published: '2026-09-24'
updated: '2026-09-24'
sourceUpdated: '2026-09-24T19:17:14.910'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-61732'
references:
  - url: >-
      https://github.com/BitterSecurity/Decepticon/commit/79ee2aaf22f4c36a5b1968f6ca3f8086b6e35b67
    label: security-advisories@github.com
  - url: >-
      https://github.com/BitterSecurity/Decepticon/security/advisories/GHSA-g5f9-3xfg-p9mf
    label: security-advisories@github.com
  - url: >-
      https://github.com/BitterSecurity/Decepticon/security/advisories/GHSA-g5f9-3xfg-p9mf
    label: 134c704f-9b21-4f2e-91b3-4a467353bcc0
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-61732'
  - url: 'https://github.com/BitterSecurity/Decepticon/pull/715'
  - url: 'https://github.com/BitterSecurity/Decepticon/releases/tag/v1.1.17'
  - url: 'https://github.com/advisories/GHSA-g5f9-3xfg-p9mf'
  - url: 'https://github.com/PurpleAILAB/Decepticon'
tags:
  - nvd
  - ghsa
  - pip
  - cve.org
  - exploit-available
  - osv
aliases:
  - GHSA-g5f9-3xfg-p9mf
ecosystem: pip
exploitAvailable: true
ssvc:
  exploitation: poc
  automatable: 'yes'
  technicalImpact: total
  timestamp: '2026-09-24T17:59:08.643777Z'
ingestedAt: '2026-09-24T17:48:30.400Z'
---

## Overview

Decepticon is an autonomous hacking agent for red teams. Versions prior to 1.1.17 wrap web crawl results — the output of agent reconnaissance against target services — into LLM messages without neutralizing ChatML special-token literals. Under the BYOK (Bring Your Own Key) deployment model, users configure their own LLM credentials to any OpenAI-compatible endpoint. Most open-source and self-deployed model providers (vLLM, SGLang, Ollama, LM Studio, text-generation-webui, etc.) do not filter special-token literals from user content in their default configurations. Those literals are parsed into structural role-boundary token IDs, meaning an attacker string planted in a target web page forges a new operator turn the model treats as authoritative, bypassing Decepticon's agent guardrails and resulting in arbitrary command execution inside the Kali Linux sandbox. Version 1.1.17 patches the issue.

## Remediation

Refer to the linked advisories for vendor-supplied fixes and affected version ranges.

## Package advisory (CVE-2026-61732)

Affected packages:

- `decepticon-core <= 1.1.16`
- `decepticon <= 1.1.16`
- `decepticon-sdk <= 1.1.16`

Patched in:

- `decepticon-core 1.1.17`
- `decepticon 1.1.17`
- `decepticon-sdk 1.1.17`

Source: https://github.com/advisories/GHSA-g5f9-3xfg-p9mf