# Defusing Explosive Prompts: Understanding and Preventing Trigger-Based Prompt Injections in LLM Agents

arXiv: https://arxiv.org/abs/2609.22510 · HTML: https://mlsanity.com/paper/2609.22510/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 30 of its 46 references that are in the map.
- First posted on arXiv 18 Sep 2026.

## Placed in

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

## Builds on (most cited first, 30 of 30)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [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)
- [BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain](https://mlsanity.com/paper/1708.06733/) (2017)
- [MPNet: Masked and Permuted Pre-training for Language Understanding](https://mlsanity.com/paper/2004.09297/) (2020)
- [Enhancing Chat Language Models by Scaling High-quality Instructional Conversations](https://mlsanity.com/paper/2305.14233/) (2023)
- [Ignore Previous Prompt: Attack Techniques For Language Models](https://mlsanity.com/paper/2211.09527/) (2022)
- [OpenAssistant Conversations -- Democratizing Large Language Model Alignment](https://mlsanity.com/paper/2304.07327/) (2023)
- [C-Pack: Packed Resources For General Chinese Embeddings](https://mlsanity.com/paper/2309.07597/) (2023)
- [InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents](https://mlsanity.com/paper/2403.02691/) (2024)
- [Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training](https://mlsanity.com/paper/2401.05566/) (2024)
- [The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions](https://mlsanity.com/paper/2404.13208/) (2024)
- [AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases](https://mlsanity.com/paper/2407.12784/) (2024)
- [StruQ: Defending Against Prompt Injection with Structured Queries](https://mlsanity.com/paper/2402.06363/) (2024)
- [Formalizing and Benchmarking Prompt Injection Attacks and Defenses](https://mlsanity.com/paper/2310.12815/) (2023)
- [Efficient Attentions for Long Document Summarization](https://mlsanity.com/paper/2104.02112/) (2021)
- [PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Models](https://mlsanity.com/paper/2402.07867/) (2024)
- [SecAlign: Defending Against Prompt Injection with Preference Optimization](https://mlsanity.com/paper/2410.05451/) (2024)
- [DataSentinel: A Game-Theoretic Detection of Prompt Injection Attacks](https://mlsanity.com/paper/2504.11358/) (2025)
- [LlamaFirewall: An open source guardrail system for building secure AI agents](https://mlsanity.com/paper/2505.03574/) (2025)
- [Adaptive Attacks Break Defenses Against Indirect Prompt Injection Attacks on LLM Agents](https://mlsanity.com/paper/2503.00061/) (2025)
- [Attention Tracker: Detecting Prompt Injection Attacks in LLMs](https://mlsanity.com/paper/2411.00348/) (2024)
- [IPIGuard: A Novel Tool Dependency Graph-Based Defense Against Indirect Prompt Injection in LLM Agents](https://mlsanity.com/paper/2508.15310/) (2025)
- [EchoLeak: The First Real-World Zero-Click Prompt Injection Exploit in a Production LLM System](https://mlsanity.com/paper/2509.10540/) (2025)
- [Breaking ReAct Agents: Foot-in-the-Door Attack Will Get You In](https://mlsanity.com/paper/2410.16950/) (2024)
- [Invitation Is All You Need! Promptware Attacks Against LLM-Powered Assistants in Production Are Practical and Dangerous](https://mlsanity.com/paper/2508.12175/) (2025)

## Abstract (arXiv)

As LLM applications integrate with external tools, they are increasingly exposed to indirect prompt injection (IPI), where adversarial instructions are embedded in retrieved content. Conventional IPIs fire on contact: the moment an agent ingests the content, it carries out the instruction. We introduce the explosive prompt, a conditional payload that stays dormant until an attacker-chosen trigger is met, in effect a training-free, inference-time backdoor planted in a single piece of retrieved content. This temporal separation reaches where ordinary IPI cannot. On frontier models that refuse the bare imperative almost entirely, rephrasing the same goal as a dormant conditional drives real, state-changing tool execution against a live agent backend (a paired mean of 16.5% vs. 2.4% for the imperative, reaching 34.2% on a proprietary model). In trials on nine production agents (OpenAI Codex, Google Gemini CLI, Anthropic Claude Code CLI, Cursor CLI, GitHub Copilot, Devin AI CLI, Amazon Kiro CLI, Qwen Code, Google Assistant; n=30 each), explosive prompts succeed in 43-83% of cases versus at most 3% for an imperative baseline, and they slip past deployed defenses: off-the-shelf injection classifiers are miscalibrated on them, and a preference-optimized model that closes imperative injection entirely still executes 11.8% of explosive prompts, every one at the trigger turn. The durable defensive lever is ingestion-time detection of the conditional structure, once detectors are trained on explosive-prompt data, which no prior benchmark supplied and our generator does. Retraining cuts live tool-execution attack success from an undefended 34.3% to 7.5-8.1% for the encoder baselines. Our detector, DeFuse, reaches 3.0% at a calibrated 5% false-positive budget with the best detection quality of any method tested (AUC 0.9994) and 25x lower latency, though it needs length-aware thresholds.