Skip to content

ActGuard: Pre-execution Action Auditing against Indirect Prompt Injection in LLM Agents

Sep 2026 · 1 citation · 72 references
Computer Science

TL;DR

Results show that ActGuard reduces attack success rates to a level comparable to state-of-the-art defenses while maintaining task utility close to the no-attack setting, achieving a favorable security-utility trade-off.

Abstract

Large language model (LLM) agents interact with external environments through tool invocation, but tool outputs can also expose them to indirect prompt injection (IPI) attacks. Existing defenses mainly rely on prompt hardening, content filtering, pre-generated plans, or permission constraints. These approaches often struggle with complex tasks or over-sanitize external content, making it difficult to balance security and utility. The key challenge is therefore to preserve execution flexibility while precisely identifying and removing the malicious content that actually induces unsafe actions. To address this challenge, we propose ActGuard, a pre-execution action auditing framework. Rather than judging whether external content is inherently suspicious, ActGuard assesses whether it causes the current action to deviate from a locally reasonable expectation. At each step, ActGuard predicts the tools likely to be used by the upcoming action and constructs a local tool prior without constraining the execution trajectory. Before execution, it compares the candidate action against this prior and performs tool-level contrastive analysis and parameter-level evidence localization to identify deviations in tool selection and action parameters. A verifier then examines the localized evidence, masks only spans confirmed as malicious, and regenerates the action from the sanitized context. This design preserves legitimate planning flexibility while minimizing information loss from indiscriminate filtering. We evaluate ActGuard on challenging benchmarks for tool-using agents. Results show that ActGuard reduces attack success rates to a level comparable to state-of-the-art defenses while maintaining task utility close to the no-attack setting, achieving a favorable security-utility trade-off. Our code is publicly available at: https://github.com/binzhwang/ActGuard.

View source

Similar papers

#artificial intelligence Preprint Oct 2026

APEX: Active Protection at Execution Boundaries for LLM Agents

APEX is an active defense that enforces both at this boundary from a single authorization contract compiled before untrusted execution, and applies uniformly across capability units without attack-specific policies or taint tracking.

Xin-Ran Zheng, Xin-Fan Guo, Zhi-Qian Hao et al. · 1 citation
Review Aug 2026

ClawSentry: A Progressive Multi-Tier Security Monitor for Safeguarding Autonomous LLM Agents

This work argues that agentic risk is progressive: it can enter at four loci of the agent control loop--skill admission, invocation-time intent, execution-time effect, and post-action consequence--while a denied dangerous objective can reappear across surface forms, tools, or turns.

Kai Wang, Zeming Wei, Biaojie Zeng et al. · 0 citations
Preprint Aug 2026

ROPE: Routed Origin Policy Enforcement against Indirect Prompt Injection

ROPE is presented, which is anchored in a structural notion of trust, and holds attack success rate to 1.6--2.6 while retaining 82--100% of undefended clean utility, significantly exceeding state-of-the-art system-level defenses in utility while attaining comparable or better security.

Xinhang Ma, Chao-Wei Xiao, William Yeoh et al. · 3 citations · ⚡2
#artificial intelligence Preprint Sep 2026

SEAD: A State-Based Perspective on Attack and Defense in Tool-Using Agents

This work forms attack and defense as partially observed state control in SEAD, deriving their design requirements from this shared execution process and constructs an environment-verifiable dataset integrating controlled initial states, replayable tool environments, and task-specific executable checks.

Xin-Jie Shen, Jun-Ran Wang, Rong-Zhe Wei et al. · 0 citations
#artificial intelligence Review Sep 2026

ActionGuard: Tool Call Authorization under Poisoned Skills

Results show that execution-boundary authorization grounded in trusted user intent and runtime evidence can restrict unauthorized Tool Calls induced by skill injection, while maintaining high benign-task completion.

Ji-Hun Han, Ye-Jin Jang, Byung Il Kwak et al. · 0 citations
Preprint Aug 2026

Reachability-Based Capability Confinement for LLM Agents under Indirect Prompt Injection

SkillGuard is presented, a harness-level enforcement layer that treats this event as contamination and restricts future capabilities to disconnect the resulting state from deployer-defined forbidden states and preserves substantially more capabilities than binary restriction at the same attack success rate.

Wu-Jie Xiong, Rabimba Karanjai, Yang Lu et al. · 2 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.