Home

Julius - LLM Service Fingerprinting Tool

Julius is an LLM service fingerprinting tool for security professionals. It detects which AI server software is running on network endpoints during penetration tests, attack surface discovery, and security assessments.

Unlike model fingerprinting tools that identify which LLM generated text, Julius identifies the server infrastructure: Is that endpoint running Ollama? vLLM? LiteLLM? A Hugging Face deployment? Julius answers in seconds.

Key Features

Feature	Description
32 LLM Services	Detects Ollama, vLLM, LiteLLM, LocalAI, Hugging Face TGI, and 27 more
Fast Scanning	Concurrent probing with intelligent port-based prioritization
Model Discovery	Extracts available models from identified endpoints via JQ expressions
Specificity Scoring	1-100 scoring ranks results by most specific match
Multiple Inputs	Single target, file input, or stdin piping
Flexible Output	Table, JSON, or JSONL formats
Extensible	Add new service detection via simple YAML probe files
Offline Operation	No cloud dependencies -- runs entirely locally
Single Binary	Go-based tool compiles to one portable executable

Quick Start

# Install
go install github.com/praetorian-inc/julius/cmd/julius@latest

# Scan a target
julius probe https://target.example.com

# List available probes
julius list

# Validate probe files
julius validate ./probes

Wiki Contents

Page	Description
Supported Services	32 LLM service probes organized by category
Probe YAML Reference	Complete probe file structure and field documentation
Match Rules	6 match rule types for response fingerprinting
CLI Reference	Complete command and flag documentation
Architecture	Project structure, data flow, and internals

Links

• GitHub Repository
• README
• Contributing Guide
• Security Policy

---

Maintained by: Praetorian