# How Adversarial Environments Mislead Agentic AI?

arXiv: https://arxiv.org/abs/2604.18874 · HTML: https://mlsanity.com/paper/2604.18874/

## Key facts

- Cited by 5 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 25 of its 42 references that are in the map.
- First posted on arXiv 20 Apr 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, 25 of 25)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs](https://mlsanity.com/paper/2307.16789/) (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)
- [Gorilla: Large Language Model Connected with Massive APIs](https://mlsanity.com/paper/2305.15334/) (2023)
- [GAIA: a benchmark for General AI Assistants](https://mlsanity.com/paper/2311.12983/) (2023)
- [Ignore Previous Prompt: Attack Techniques For Language Models](https://mlsanity.com/paper/2211.09527/) (2022)
- [Baseline Defenses for Adversarial Attacks Against Aligned Language Models](https://mlsanity.com/paper/2309.00614/) (2023)
- [How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs](https://mlsanity.com/paper/2401.06373/) (2024)
- [Identifying the Risks of LM Agents with an LM-Emulated Sandbox](https://mlsanity.com/paper/2309.15817/) (2023)
- [Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflicts](https://mlsanity.com/paper/2305.13300/) (2023)
- [Defending Against Indirect Prompt Injection Attacks With Spotlighting](https://mlsanity.com/paper/2403.14720/) (2024)
- [Certifiably Robust RAG against Retrieval Corruption](https://mlsanity.com/paper/2405.15556/) (2024)
- [AVeriTeC: A Dataset for Real-world Claim Verification with Evidence from the Web](https://mlsanity.com/paper/2305.13117/) (2023)
- [The Semantic Scholar Open Data Platform](https://mlsanity.com/paper/2301.10140/) (2023)
- [Do Language Models Know When They're Hallucinating References?](https://mlsanity.com/paper/2305.18248/) (2023)
- [TrustRAG: Enhancing Robustness and Trustworthiness in Retrieval-Augmented Generation](https://mlsanity.com/paper/2501.00879/) (2025)
- [ReportBench: Evaluating Deep Research Agents via Academic Survey Tasks](https://mlsanity.com/paper/2508.15804/) (2025)
- [Attractive Metadata Attack: Inducing LLM Agents to Invoke Malicious Tools](https://mlsanity.com/paper/2508.02110/) (2025)
- [ADVEDM:Fine-grained Adversarial Attack against VLM-based Embodied Agents](https://mlsanity.com/paper/2509.16645/) (2025)
- [RAG Security and Privacy: Formalizing the Threat Model and Attack Surface](https://mlsanity.com/paper/2509.20324/) (2025)

## Built on by (most cited first, 5 of 5)

- [MetaResearcher: Scaling Deep Research via Self-Reflective Reinforcement Learning in Adversarial Virtual Environments](https://mlsanity.com/paper/2606.19893/) (2026)
- [Labels Are Not Endpoints: Treatment Leakage and Construct Validity in MCP Agent Security Evaluation](https://mlsanity.com/paper/2608.12880/) (2026)
- [Training-Free Inference-Time Self-Reflection and Cost-Bounded Early Stopping for Large Language Models](https://mlsanity.com/paper/2608.18884/) (2026)
- [Record Grouping Controls Evidence Weight in Language Models](https://mlsanity.com/paper/2609.08698/) (2026)
- [Can agents whose sole duty is to keep humans alive be hijacked to kill? Delayed multi-turn derived-action trap attacks in simulated high-stakes environments](https://mlsanity.com/paper/2610.15153/) (2026)

## Abstract (arXiv)

Tool-integrated agents are deployed on the premise that external tools ground their outputs in reality. Yet this very reliance creates a critical attack surface. Current evaluations benchmark capability in benign settings, asking "can the agent use tools correctly" but never "what if the tools lie". We identify this Trust Gap: agents are evaluated for performance, not for skepticism. We formalize this vulnerability as Adversarial Environmental Injection (AEI), a threat model where adversaries compromise tool outputs to deceive agents. AEI constitutes environmental deception: constructing a "fake world" of poisoned search results and fabricated reference networks around unsuspecting agents. We operationalize this via POTEMKIN, a Model Context Protocol (MCP)-compatible harness for plug-and-play robustness testing. We identify two orthogonal attack surfaces: The Illusion (breadth attacks) poison retrieval to induce epistemic drift toward false beliefs, while The Maze (depth attacks) exploit structural traps to cause policy collapse into infinite loops. Across 11,000+ runs on five frontier agents, we find a stark robustness gap: resistance to one attack often increases vulnerability to the other, demonstrating that epistemic and navigational robustness are distinct capabilities.