Text Generation
Transformers
English
security
pentesting
autonomous-agent
cybersecurity
tool-use
qwen2.5
Instructions to use automajicly/Local_Security_Model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use automajicly/Local_Security_Model with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="automajicly/Local_Security_Model")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("automajicly/Local_Security_Model", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use automajicly/Local_Security_Model with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "automajicly/Local_Security_Model" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "automajicly/Local_Security_Model", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/automajicly/Local_Security_Model
- SGLang
How to use automajicly/Local_Security_Model with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "automajicly/Local_Security_Model" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "automajicly/Local_Security_Model", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "automajicly/Local_Security_Model" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "automajicly/Local_Security_Model", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use automajicly/Local_Security_Model with Docker Model Runner:
docker model run hf.co/automajicly/Local_Security_Model
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,93 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- security
|
| 5 |
+
- pentesting
|
| 6 |
+
- autonomous-agent
|
| 7 |
+
- cybersecurity
|
| 8 |
+
- tool-use
|
| 9 |
+
- qwen2.5
|
| 10 |
+
language:
|
| 11 |
+
- en
|
| 12 |
+
base_model:
|
| 13 |
+
- bartowski/Qwen2.5-14B_Uncensored_Instruct-GGUF
|
| 14 |
+
pipeline_tag: text-generation
|
| 15 |
+
library_name: transformers
|
| 16 |
---
|
| 17 |
+
|
| 18 |
+
π Local Security Model β Autonomous Pentesting Agent
|
| 19 |
+
|
| 20 |
+
Developed by: automajicly
|
| 21 |
+
Built on: Qwen2.5-14b-Instruct-Uncensored-GGUF by Bartowski
|
| 22 |
+
|
| 23 |
+
OVERVIEW
|
| 24 |
+
|
| 25 |
+
Local_Security_Model is an autonomous penetration testing agent designed for professional security assessments. Built on top of Qwen 2.5, it operates through a custom MCP (Model Context Protocol) architecture that enables real-time tool orchestration, vulnerability discovery, and exploit chaining β all running locally with no cloud dependency.
|
| 26 |
+
This agent was developed as the core engine behind PenMaster Security, targeting small business security audits, WordPress hardening, and ecommerce vulnerability assessments.
|
| 27 |
+
|
| 28 |
+
Key Capabilities
|
| 29 |
+
β’ Autonomous reconnaissance β masscan + nmap port/service enumeration with zero manual input
|
| 30 |
+
β’ Vulnerability assessment β searchsploit integration for CVE matching against discovered services
|
| 31 |
+
β’ Web application testing β nikto and sqlmap for injection and misconfiguration detection
|
| 32 |
+
β’ Credential auditing β hydra and ncrack for multi-protocol brute force testing
|
| 33 |
+
β’ Payload delivery β curl/wget for staged payload retrieval and HTTP interaction
|
| 34 |
+
β’ Structured reporting β auto-generated HTML pentest reports with severity ratings and remediation guidance
|
| 35 |
+
|
| 36 |
+
Architecture
|
| 37 |
+
|
| 38 |
+
agent_loop.py β LLM reasoning + tool chain generation (Qwen 2.5 via LM Studio)
|
| 39 |
+
mcp_server.py β Flask-based tool execution server (port 8000, systemd managed)
|
| 40 |
+
report_generator.py β HTML report engine with PenMaster branding
|
| 41 |
+
logs/ β Structured JSON session logs
|
| 42 |
+
reports/ β Auto-generated client-facing pentest reports
|
| 43 |
+
|
| 44 |
+
|
| 45 |
+
Model backend:
|
| 46 |
+
Qwen 2.5-14B served locally via LM StudioExecution layer: Flask MCP server with systemd auto-restartInterface: Terminal-native, SSH-accessible from remote IDEs (Cursor)
|
| 47 |
+
|
| 48 |
+
Tool Stack:
|
| 49 |
+
|
| 50 |
+
TOOL PURPOSE
|
| 51 |
+
|
| 52 |
+
masscan High speed port scanning
|
| 53 |
+
|
| 54 |
+
nmap Service/version enumeration
|
| 55 |
+
|
| 56 |
+
nitko Web server vulnerability scanning
|
| 57 |
+
|
| 58 |
+
sqlmap SQL injection detection
|
| 59 |
+
|
| 60 |
+
hydra Multi-protocol credential brute forcing
|
| 61 |
+
|
| 62 |
+
ncrack Network authentication cracking
|
| 63 |
+
|
| 64 |
+
searchsploit CVE/exploit database lookup
|
| 65 |
+
|
| 66 |
+
curl/wget HTTP interaction and payload staging
|
| 67 |
+
|
| 68 |
+
|
| 69 |
+
|
| 70 |
+
Intended Use
|
| 71 |
+
|
| 72 |
+
This model and agent stack is designed for:
|
| 73 |
+
β’ Professional penetration testing against authorized targets
|
| 74 |
+
β’ Security audits for small businesses, WordPress sites, and ecommerce platforms
|
| 75 |
+
β’ Vulnerability research in isolated lab environments
|
| 76 |
+
β’ Security education and CTF preparation
|
| 77 |
+
β οΈ Authorized use only. This tool is intended exclusively for use against systems you own or have explicit written permission to test. Unauthorized use is illegal and unethical.
|
| 78 |
+
|
| 79 |
+
|
| 80 |
+
Target Environments
|
| 81 |
+
β’ Kali Linux (primary deployment platform)
|
| 82 |
+
β’ Isolated VM lab networks
|
| 83 |
+
β’ Small business web infrastructure (with client authorization)
|
| 84 |
+
|
| 85 |
+
|
| 86 |
+
Business Context
|
| 87 |
+
Local_Security_Model is the core engine behind PenMaster Security β an independent penetration testing project offering:
|
| 88 |
+
β’ Initial security audit and vulnerability report
|
| 89 |
+
β’ Ongoing security hardening retainer
|
| 90 |
+
β’ WordPress and ecommerce-focused assessments
|
| 91 |
+
π¬ Contact: GitHub.com/XenoCoreGiger31
|
| 92 |
+
|
| 93 |
+
|