# Securing AI agents and tool calls

DevFeed: [Securing AI agents and tool calls](<https://devfeed.tech/articles/securing-ai-agents-and-tool-calls-16644.md>)

Original publisher: [Read original article](<https://firebase.blog/posts/2025/12/securing-ai-agents>)

Author: Alexander Nohe

Published: 2025-12-02T19:00:00Z

Content type: tutorial

Language: en

Sources: [Firebase Blog](<https://devfeed.tech/sources/firebase-blog.md>)

Topics: [Securing AI](<https://devfeed.tech/topics/securing-ai.md>), [prompt injection](<https://devfeed.tech/topics/prompt-injection.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Loop Engineering](<https://devfeed.tech/topics/loop-engineering.md>), [App](<https://devfeed.tech/topics/app.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [app-check](<https://devfeed.tech/tags/app-check.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [firebase](<https://devfeed.tech/tags/firebase.md>), [genkit](<https://devfeed.tech/tags/genkit.md>), [go](<https://devfeed.tech/tags/go.md>), [least-privilege](<https://devfeed.tech/tags/least-privilege.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [prompt-injection](<https://devfeed.tech/tags/prompt-injection.md>), [securing-ai](<https://devfeed.tech/tags/securing-ai.md>)

## AI overview

This article explains how prompt injection can manipulate AI agent tool calls to access unauthorized user data. It recommends using application context and least-privilege access when designing tool definitions.

## Source excerpt

Beyond authentication: Learn how to secure your AI agent's tool calls from prompt injection by using application context and the principle of least privilege.