KadirArslan/Mithra-Scanner

Mithra Scanner is an interactive API testing tool for prompt injection, refusal detection, and LLM security benchmarking. It supports YAML-based rule definitions, custom refusal lists, REST API integration, and provides detailed CLI output for security testing of language model endpoints.

17
/ 100
Experimental

This tool helps security engineers and AI product managers test their large language model (LLM) APIs for vulnerabilities like prompt injection and unwanted content generation. It takes a list of predefined attack prompts and refusal phrases, feeds them into your LLM endpoint, and reports which attacks succeeded or were properly refused. This allows you to benchmark your LLM's security against known threats.

No commits in the last 6 months.

Use this if you need to systematically test your LLM's security and ensure it doesn't respond to malicious prompts or generate undesirable content.

Not ideal if you are looking for a general-purpose LLM evaluation framework or a tool to analyze LLM performance on standard benchmarks.

LLM security prompt injection API security testing AI risk management red teaming
No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 8 / 25
Maturity 7 / 25
Community 0 / 25

How are scores calculated?

Stars

52

Forks

Language

Python

License

Last pushed

Sep 09, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/KadirArslan/Mithra-Scanner"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.