Browse categories
Explore
Fiverr Pro
English
$
USD
I will audit your ai system for prompt injection, data poisoning and leakage
As a CyberSecurity student with interest in reading, grammar, and tech
As a CyberSecurity student, Im also interested in reading and writing about english topics and technology. Its one of my many passions. I want to read as much as I can and help others in their content...
About this Gig
Is your AI app safe from prompt injection?
Most LLM chatbots and AI features ship without ever being tested against an attacker. A single crafted input can override your system prompt, leak your configuration, or expose customer data.
I'm a cybersecurity student and active bug bounty hunter specializing in LLM security. I test your AI application the way a real attacker would.
Basic package covers:
- Direct and indirect prompt injection testing
- System prompt extraction and configuration leakage
- Sensitive information disclosure (PII, keys, internal data in model output)
- Data poisoning exposure review across your training and RAG pipeline
- PDF report with severity ratings, reproduction steps, and fix guidance
You will get a clear report your developers can act on immediately not a tool dump.
Works with OpenAI, Anthropic, Gemini, open-source models, and custom RAG setups.
Written authorization from the system owner is required before testing begins.
Message me before ordering with your app type and I'll confirm scope and turnaround.
