Metraipolis › archive › dc_ce94b0aed16f

ActGuard: Pre-execution Action Auditing against Indirect Prompt Injection in LLM Agents — arXiv:2609.14987

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.14987
License (as archived)
CC0-1.0
Archived
2026-10-15T22:53:29.448Z
SHA-256 of the archived text
0d1da25b0e810668221daa9dddbf95332942a09ebd251964ba203fd7ec2a4701
Stable URL of this page
https://metraipolis.com/documento/dc_ce94b0aed16f

How to cite

ActGuard: Pre-execution Action Auditing against Indirect Prompt Injection in LLM Agents — arXiv:2609.14987 — arXiv (en), https://arxiv.org/abs/2609.14987 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ce94b0aed16f (Metraipolis archive dc_ce94b0aed16f, sha256 0d1da25b0e810668221daa9dddbf95332942a09ebd251964ba203fd7ec2a4701)
BibTeX
@misc{dcce94b0aed16f,
  title = {ActGuard: Pre-execution Action Auditing against Indirect Prompt Injection in LLM Agents — arXiv:2609.14987},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.14987},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ce94b0aed16f; sha256 0d1da25b0e810668221daa9dddbf95332942a09ebd251964ba203fd7ec2a4701},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.14987. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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: this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.14987 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T22:29:03.766Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).