Prompt Injection Defense with Python: Protect LLM Apps, RAG Pipelines, AI Agents, and Tool-Calling Workflows from Real-World Prompt Attacks - Softcover

Calado, Leandro

9798196154553: Prompt Injection Defense with Python: Protect LLM Apps, RAG Pipelines, AI Agents, and Tool-Calling Workflows from Real-World Prompt Attacks

Softcover

ISBN 13: 9798196154553

Publisher: Independently published, 2026

View all copies of this ISBN edition

0 Used

1 New

From US$ 68.00

Protect your AI systems from prompt injection attacks before they reach production.

Most developers are building LLM apps, RAG pipelines, and AI agents without a real security layer. This book shows you how to fix that with practical Python projects you can build, test, deploy, and turn into paid services.

Prompt Injection Defense with Python is a hands-on guide for developers, AI engineers, freelancers, and technical founders who want to secure modern LLM applications using Python 3.11, FastAPI, ChromaDB, SQLite, and Docker.

Inside, you will build practical AI security projects such as:

Prompt Firewall API — a FastAPI middleware that detects risky prompts, blocks malicious inputs, and assigns threat scores.
Secure RAG Pipeline — document ingestion with malicious content detection, retrieval validation, and safer context handling.
Agent Approval Gateway — a permission and audit system for AI agents that call sensitive tools and APIs.
Prompt Injection Scanner — a testing library with attack payloads and JSON vulnerability reports.
Security Dashboard — a lightweight monitoring system for attacks, logs, metrics, and incidents using SQLite.

This book focuses on real implementation, not theory. You will learn how to design defensive layers around LLM applications, audit vulnerable workflows, monitor suspicious behavior, and create reusable tools that can become portfolio projects, consulting offers, or MicroSaaS products.

You will also learn how to package your skills into paid AI security services, including LLM app audits, RAG security reviews, agent risk assessments, and subscription-based security tooling.

If you want to build safer AI applications and turn LLM security into a practical business opportunity, this book gives you the projects, architecture, and code patterns to start.

"synopsis" may belong to another edition of this title.