# Abusing Function Calling for Unauthorized Actions

> Master how attackers manipulate LLM tool arguments to perform unauthorized actions, escalate privileges, and bypass access controls in agentic workflows.

Canonical: https://codelivly.com/rooms/abusing-function-calling-for-unauthorized-actions
Last updated: 2026-10-12

| | |
|---|---|
| Level | Intermediate |
| Category | Cybersecurity |
| Lessons | 3 |
| Estimated time | 30 min |
| XP | 100 |

This module explores the attacker's playbook for exploiting function calling. You will learn how to manipulate tool arguments to trigger Insecure Direct Object References (IDOR), escalate privileges, bypass business logic, and execute unauthorized actions by tricking the LLM's routing and parameter extraction.

## Lessons

1. What is Function-Calling Abuse (30 min)
2. Tricking a Model into Calling Unauthorized Functions (30 min)
3. Chaining Multiple Tool Calls for Privilege Escalation (30 min)

## Related

- [Learning path: AI API and Function-Calling Attacks](https://codelivly.com/learning-paths/ai-api-and-function-calling-attacks)
- [All rooms](https://codelivly.com/rooms)