# Blind Spots in the Guard: How Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems

arXiv: https://arxiv.org/abs/2605.22001 · HTML: https://mlsanity.com/paper/2605.22001/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 7 of its 11 references that are in the map.
- First posted on arXiv 21 May 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 7 of 7)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection](https://mlsanity.com/paper/2302.12173/) (2023)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate](https://mlsanity.com/paper/2305.19118/) (2023)
- [Ignore Previous Prompt: Attack Techniques For Language Models](https://mlsanity.com/paper/2211.09527/) (2022)

## Built on by (most cited first, 3 of 3)

- [SoK: When Safe Agents Fail Together: The Security of Multi Agent LLM Systems](https://mlsanity.com/paper/2609.00595/) (2026)
- [PARSE: Provenance-Aware Retrieval Sanitization for Professional Domain LLM Agents](https://mlsanity.com/paper/2606.17467/) (2026)
- [Evaluating Prompting-Based Defenses Against Domain-Camouflaged Injection Attacks](https://mlsanity.com/paper/2606.18530/) (2026)

## Abstract (arXiv)

Injection detectors deployed to protect LLM agents are calibrated on static, template-based payloads that announce themselves as override directives. We identify a systematic blind spot: when payloads are generated to mimic the domain vocabulary and authority structures of the target document, what we call domain camouflaged injection, standard detectors fail to flag them, with detection rates dropping from 93.8% to 9.7% on Llama 3.1 8B and from 100% to 55.6% on Gemini 2.0 Flash. We formalize this as the Camouflage Detection Gap (CDG), the difference in injection detection rate between static and camouflaged payloads. Across 45 tasks spanning three domains and two model families, CDG is large and statistically significant (chi^2 = 38.03, p < 0.001 for Llama; chi^2 = 17.05, p < 0.001 for Gemini), with zero reverse discordant pairs in either case. We additionally evaluate Llama Guard 3, a production safety classifier, which detects zero camouflage payloads (IDRcamouflage = 0.000), confirming that the blind spot extends beyond few-shot detectors to dedicated safety classifiers. We further show that multi-agent debate architectures amplify static injection attacks by up to 9.9x on smaller models, while stronger models show collective resistance. Targeted detector augmentation provides only partial remediation (10.2% improvement on Llama, 78.7% on Gemini), suggesting the vulnerability is architectural rather than incidental for weaker models. Our framework, task bank, and payload generator are released publicly.