Skip to content

Hide Navigation Hide TOC

Word-Fragment Concatenation Assembly Attack - ATR-2026-00443 (b8a2b41d-20b4-517c-8a72-1f2c487da764)

Detects prompts asking the model to assemble a target string from explicitly enumerated word fragments or concatenated substrings, a tokenisation-bypass shape common in prompt-hacking corpora (HackAPrompt). The attacker splits the dangerous output across multiple tokens so naive prompt-token filters miss it, then asks the model to concatenate the pieces. Captures both natural-language enumeration ("first word is X, second is Y...") and code-like concat ("X = 'foo' + 'bar' + 'baz'"). Generalises beyond PWNED: same attack shape used to assemble "kill", "explode", "wire transfer" etc.

Cluster A Galaxy A Cluster B Galaxy B Level
LLM Prompt Injection (19cd2d12-66ff-487c-a05c-e058b027efc9) MITRE ATLAS Attack Pattern Word-Fragment Concatenation Assembly Attack - ATR-2026-00443 (b8a2b41d-20b4-517c-8a72-1f2c487da764) Agent Threat Rules 1
LLM Jailbreak (172427e3-9ecc-49a3-b628-96b824cc4131) MITRE ATLAS Attack Pattern Word-Fragment Concatenation Assembly Attack - ATR-2026-00443 (b8a2b41d-20b4-517c-8a72-1f2c487da764) Agent Threat Rules 1