Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models

Yannis Belkhiter, Giulio Zizzo, Sergio Maffeis, Seshu Tirupathi +1 more

Published

Apr 22, 2026

Citations

0

Trust level

High

Usefulness score

65/100 (Medium)

Extraction confidence

80% (High)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Aug 31, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

Automatic Metrics General Pairwise Preference Red Team Tool Use

Should you rely on this paper?

This paper has useful evaluation signal, but protocol completeness is partial; pair it with related papers before deciding implementation strategy.

Use this as a practical starting point for protocol research, then validate against the original paper.

Best use

Secondary protocol comparison source

Use if you need

A benchmark-and-metrics comparison anchor.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

No major weakness surfaced.

Human feedback signal
Detected
From extracted signals
Evaluation signal
Detected
Eval setup described
Usefulness for eval research
65/100
Moderate-confidence candidate

Useful as a secondary reference; validate protocol details against neighboring papers.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions. Injection and jailbreaking attacks have been extensively explored to showcase the vulnerabilities of LLMs to user prompt manipulation. The expanded capabilities of agentic models introduce further vulnerabilities via their function calling interface. Recent work in LLM security showed that function calling can be abused, leading to data tampering and theft, causing disruptive behavior such as endless loops, or causing LLMs to produce harmful content in the style of jailbreaking attacks. This paper introduces a novel function hijacking attack (FHA) that manipulates the tool selection process of agentic models to force the invocation of an attacker-chosen function. While existing attacks focus on semantic preference of the model for function-calling tasks, we show that FHA is largely agnostic to the context semantics and remains effective across domains and function sets. We demonstrate that FHA generalizes to unseen queries and payload perturbations under a fixed target model, reaching 62.5% to 81.9% ASR on held-out queries across 4 function-calling LLMs (instructed and reasoning models), evaluated on the Berkeley Function Calling Leaderboard (BFCL). We further evaluate the cross-model transferability of FHA, showing that FHA can be transferred to other model sizes and families (11.2-27.6% ASR). Our findings further demonstrate the need for strong guardrails and modules for agentic systems.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

strong

Pairwise Preference, Red Team

Directly usable for protocol triage.

"The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions."

Evaluation Modes

strong

Automatic Metrics

Includes extracted eval setup.

"The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions."

Quality Controls

missing

Not reported

No explicit QC controls found.

"The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions."

Benchmarks / Datasets

strong

BFCL

Useful for quick benchmark comparison.

"We demonstrate that FHA generalizes to unseen queries and payload perturbations under a fixed target model, reaching 62.5% to 81.9% ASR on held-out queries across 4 function-calling LLMs (instructed and reasoning models), evaluated on the Berkeley Function Calling Leaderboard (BFCL)."

Reported Metrics

strong

Jailbreak success rate

Useful for evaluation criteria comparison.

"The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions."

Benchmarks and datasets

BFCL

Reported metrics

jailbreak success rate
Human feedback details
Uses human feedback
Yes
Feedback types
Pairwise Preference, Red Team
Rater population
Not reported
Expertise required
General
Evaluation details
Evaluation modes
Automatic Metrics
Agentic eval
Tool Use
Quality controls
Not reported
Evidence quality
High
Use this page as
Secondary protocol comparison source

Research brief

Metadata summary

The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions.
  • Injection and jailbreaking attacks have been extensively explored to showcase the vulnerabilities of LLMs to user prompt manipulation.
  • The expanded capabilities of agentic models introduce further vulnerabilities via their function calling interface.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Check the full text for explicit evaluation design choices (raters, protocol, and metrics).
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Pairwise preference evaluation

Contribution summary

  • The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions.
  • While existing attacks focus on semantic preference of the model for function-calling tasks, we show that FHA is largely agnostic to the context semantics and remains effective across domains and function sets.
  • We demonstrate that FHA generalizes to unseen queries and payload perturbations under a fixed target model, reaching 62.5% to 81.9% ASR on held-out queries across 4 function-calling LLMs (instructed and reasoning models), evaluated on the…

Why it matters for eval

  • The growth of agentic AI has drawn significant attention to function calling Large Language Models (LLMs), which are designed to extend the capabilities of AI-powered system by invoking external functions.
  • While existing attacks focus on semantic preference of the model for function-calling tasks, we show that FHA is largely agnostic to the context semantics and remains effective across domains and function sets.

Researcher checklist

  • Human feedback protocol is explicit

    Detected: Pairwise Preference, Red Team

  • Evaluation mode is explicit

    Detected: Automatic Metrics

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    Detected: BFCL

  • Metric reporting is present

    Detected: jailbreak success rate