Indirect Prompt Injection in Municipal Document-Processing Copilots: Attacks, Defences and Harm

Abstract

Public administrations are deploying large language model (LLM) assistants that process, summarise, classify and validate citizen-submitted documents. These copilots are exposed to indirect prompt injection: instructions hidden in manipulated documents that reach the model as if they were data. We develop a municipal aid-procedure copilot and evaluate its robustness across six attack objectives, five delivery vectors, five defences and four LLMs, with 3,000 attack evaluations and 1,000

Categories

Framework mappings

Suggested from the entry's categories.

@misc{cisnerosgonzalez2026indirect,
  title = {{Indirect Prompt Injection in Municipal Document-Processing Copilots: Attacks, Defences and Harm}},
  author = {Jorge Cisneros-González and José Antonio Ondiviela García and Javier Sánchez-Soriano},
  year = {2026},
  month = sep,
  doi = {10.62161/sauc.v12.6399},
  url = {https://doi.org/10.62161/sauc.v12.6399}
}