Goal Reached Thanks to every supporter — we hit 100%!

Goal: 1000 CNY · Raised: 1359 CNY

100%
Buy Us a Coffee

CVE-2026-61732— Decepticon: Role-boundary forgery via ChatML special-token literals in web crawl output composed into LLM context

Quick assessment

Affected
BitterSecurity Decepticon
Exploitation
No confirmed in-the-wild exploitation; assess based on exposure
Recommended action
Check the vendor advisory and references for a fixed version. If immediate upgrade is impossible, restrict exposure and increase monitoring.

Decepticon 是专为红队测试设计的自主黑客代理工具。在 1.1.17 版本之前，该工具在将网络爬虫结果（即代理对目标服务进行侦察时产生的输出）封装为 LLM（大语言模型）消息时，未对 ChatML 特殊令牌字面量进行转义或过滤处理。 在 BYOK（Bring Your Own Key，自带密钥）部署模式下，用户可自行配置任意与 OpenAI 兼容的 LLM 端点所使用的凭证。大多数开源模型提供商及自托管部署方案（如 vLLM、SGLang、Ollama、LM Studio、text-generation-w

CVSS 10.0 · Critical EPSS 1.22% · P67
Get alerts for future matching vulnerabilities Log in to subscribe

I. Basic Information for CVE-2026-61732

Vulnerability Information
Have questions about the vulnerability? See if Shenlong's analysis helps!
View Shenlong Deep Dive ↗

Although we use advanced large model technology, its output may still contain inaccurate or outdated information.Shenlong tries to ensure data accuracy, but please verify and judge based on the actual situation.

Vulnerability Title
Decepticon: Role-boundary forgery via ChatML special-token literals in web crawl output composed into LLM context
Source: CVE Program / CVE List V5
Vulnerability Description
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.
Source: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H
Source: CVE Program / CVE List V5
Vulnerability Type
输出中的特殊元素转义处理不恰当(注入)
Source: CVE Program / CVE List V5

Affected Products

Vendor Product Affected Versions CPE Subscribe
BitterSecurity Decepticon < 1.1.17 -
BitterSecurity decepticon-core < 1.1.17 -
BitterSecurity decepticon-sdk < 1.1.17 -

II. Public POCs for CVE-2026-61732

# POC Description Source Link Shenlong Link
AI-Generated POC Premium

No public POC found.

Login to generate AI POC

III. Intelligence Information for CVE-2026-61732

请登录查看更多情报信息。

Other References for CVE-2026-61732 (2)

IV. Related Vulnerabilities

Same weakness: CWE-74

V. Comments for CVE-2026-61732

No comments yet

Leave a comment