Right Tool, Wrong Parameters: Defending LLM Agents Against Parameter Poisoning

Summary

This paper studies parameter poisoning in LLM agents, where an indirect prompt injection keeps the intended action but manipulates a tool argument such as a recipient. The authors propose AuthorityFlow, a three-stage defense that builds a task-specific trust hierarchy from the user query, traces each argument to its supporting information field, and flags low-trust influence. On their FactAuthBench evaluation, attack success falls from 93.06% to 9.03%, with 82.64% benign task utility retained versus 90.28% without defense.

Mitigation

AuthorityFlow: a three-stage parameter-level defense that constructs a task-specific trust hierarchy over information fields, traces each tool argument to its supporting field, and accepts a value only when that field is trusted to determine the parameter under the current task.

Topics

• Prompt injection and jailbreaks

Related items

• InfoTop agent skill security resources — November 2026Similar attack · Adversa AI Blog
• InfoHow to Turn AI Red Team Findings Into Runtime ControlsSimilar attack · Straiker Blog
• MediumDocker Sandboxes vulnerability lets AI agents access sensitive files outside their sandbox (CVE-2026-103916) | Endor LabsSimilar attack · Endor Labs Blog
• MediumFree the Model, Sell the Gateway: Prompt Adware hidden in Jailbreak BundlesSimilar attack · Check Point Research
• InfoThe Devil is in the Context: Nine Dimensions of Prompt Injection AttacksSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)