CVE-2026-61732Critical· 10.0▾ AbyssalPoC availableDecepticon 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.…
▾ Abyssal zone — Critical with a public exploit or in-the-wild use
impact 55 · likelihood 0.3 · exploitation 12
A public proof-of-concept already exists for this vulnerability — see Exploit availability below.
Public exploit / PoC code seen in 1 source. Availability, not in-the-wild use.
Stakeholder-Specific Vulnerability Categorization from CISA's ADP record at CVE.org: whether exploitation is observed, whether an attack can be automated, and how much of the system is at stake.
Exploit-prediction probability, daily snapshots since Sep 26.
Disclosure to exploitation, from the record and what we observed since indexing it.
Disclosed via NVD
1 GitHub repo
1.2%
Last analysed / modified upstream
1.2% → 1.5%
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.
Refer to the linked advisories for vendor-supplied fixes and affected version ranges.
Affected packages:
decepticon-core <= 1.1.16decepticon <= 1.1.16decepticon-sdk <= 1.1.16Patched in:
decepticon-core 1.1.17decepticon 1.1.17decepticon-sdk 1.1.17Field changes observed since this record was first indexed.
Connected by shared product, vendor, weakness, or advisory.
CVE-2023-32785Critical· 9.8Langchain SQL Injection vulnerability
CVE-2023-29383Low· 3.3In Shadow 4.13, it is possible to inject control characters into fields provided to the SUID program chfn (change finger)
CVE-2023-27533High· 8.8A vulnerability in input validation exists in curl <8.0 during communication using the TELNET protocol may allow an attacker to pass on maliciously crafted user name and "telnet options" during server negotiation
CVE-2023-4548Medium· 6.3A vulnerability has been found in SPA-Cart eCommerce CMS 1.9.0.3
CVE-2025-10012Medium· 6.3A security vulnerability has been detected in Portabilis i-Educar up to 2.10
CVE-2019-2725Critical· 9.8Vulnerability in the Oracle WebLogic Server component of Oracle Fusion Middleware (subcomponent: Web Services)