Instructions to use cortexrouter/judge-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use cortexrouter/judge-v1 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B") model = PeftModel.from_pretrained(base_model, "cortexrouter/judge-v1") - Notebooks
- Google Colab
- Kaggle
CortexRouter Judge v1
Apache-2.0 Open-Weights Vulnerability-Evidence Verification Model (Qwen3-8B LoRA)
"FIND IT BEFORE THEY DO." โ Offline Air-Gapped Security Triage & 9-Field Decision Reporting
Executive Overview
CortexRouter Judge v1 (cortexrouter/judge-v1) is a domain-specialized PEFT / LoRA adapter built on Qwen/Qwen3-8B for automated vulnerability-evidence verification and false-positive elimination in authorized security testing and Coordinated Vulnerability Disclosure (CVD) workflows.
Traditional web scanners flag HTTP 200 OK responses, generic login pages, or custom 404/500 error templates as critical CVE findings. CortexRouter Judge v1 runs 100% offline on local air-gapped GPUs (zero source code or sensitive crash trace leakage to external cloud APIs), inspects the raw HTTP/CLI test evidence alongside target metadata, and emits a deterministic 9-field structured triage report.
Architecture & Triage Pipeline
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ ISOLATED LAB / SCANNER โ โ CORTEXROUTER JUDGE v1 โ โ 9-FIELD STRUCTURED REPORT โ
โ โข Target & Version Metadata โ โโโโถ โ โข Base: Qwen/Qwen3-8B (bfloat16) โ โโโโถ โ โข DOฤRULANDI (Confirmed) โ
โ โข Raw HTTP / CLI Evidence โ โ โข Adapter: LoRA (Apache-2.0) โ โ โข ETKฤฐSฤฐZ (Patched / Closed) โ
โ โข Candidate CVE Class โ โ โข 100% Local / Air-Gapped vLLM โ โ โข YANLIล_POZฤฐTฤฐF (False Pos.) โ
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
Structured 9-Field Output Schema
Given raw security test evidence, the model outputs a strict 9-field structured decision report:
- Hedef ortam (Target environment)
- Zafiyet kimliฤi (Vulnerability / CVE ID)
- Kategori (Vulnerability category / CWE class)
- Test adฤฑmlarฤฑ (Verification steps)
- Gรถzlemlenen kanฤฑt (Observed technical evidence)
- Karar (Final triage decision:
DOฤRULANDI|ETKฤฐSฤฐZ|YANLIล_POZฤฐTฤฐF) - Gerekรงe (Evidence-backed technical rationale)
- Gรผven dรผzeyi (Confidence assessment)
- Kaynak (Reference / advisory context)
| Decision Class | Meaning | Operational Action |
|---|---|---|
DOฤRULANDI (Confirmed) |
Concrete exploitability / impact artifact is present in the observed output (e.g., /etc/passwd leak, SQL syntax trace, command execution proof). |
Escalate to human researcher & initiate 90-day Coordinated Vulnerability Disclosure. |
ETKฤฐSฤฐZ (Ineffective / Patched) |
Vendor patch applied (HTTP 404), vulnerable component disabled (HTTP 500), or endpoint unreachable. |
Close as mitigated / non-reproducible in target configuration. |
YANLIล_POZฤฐTฤฐF (False Positive) |
Request returned HTTP 200 (e.g., redirect to standard login page or static HTML) with zero security impact. |
Suppress scanner noise automatically. |
Language Design Note: Output field labels and decision classes are intentionally standardized in Turkish (
DOฤRULANDI,ETKฤฐSฤฐZ,YANLIล_POZฤฐTฤฐF) for deterministic downstream parsing (eval/run_eval.py). Both Turkish and English evidence prompts are understood.
Verified Holdout Benchmark (80/80 = 100%)
Evaluated on cortexrouter/judge-v1-holdout โ a clean, unseen holdout benchmark spanning 20 distinct CVEs ร 4 adversarial triage scenarios = 80 test cases:
| Scenario | Test Condition | Expected Decision | Accuracy |
|---|---|---|---|
| Scenario A | Concrete exploit evidence present in output | DOฤRULANDI |
20/20 (100%) |
| Scenario B | Vendor-patched target (HTTP 404 / error page, no impact) |
ETKฤฐSฤฐZ |
20/20 (100%) |
| Scenario C | Vulnerable version but required module/component disabled | ETKฤฐSฤฐZ |
20/20 (100%) |
| Scenario D | Ambiguous HTTP 200 login/landing page flagged by scanner |
YANLIล_POZฤฐTฤฐF |
20/20 (100%) |
| Total | 20 Unseen CVEs ร 4 Scenarios | All 3 Classes | 80/80 (100.0%) |
Reproduce the Benchmark Yourself
# Clone model or GitHub evaluation suite
git clone https://github.com/CortexRouter/cortexrouter.git
cd cortexrouter
# Option 1: Evaluate directly with Transformers + PEFT
python eval/run_eval.py --model cortexrouter/judge-v1
# Option 2: Evaluate against a local vLLM LoRA server
python eval/run_eval.py --api http://localhost:8000/v1 --model judge
Quickstart
1. Python (transformers + peft)
pip install transformers peft accelerate torch
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
BASE_MODEL = "Qwen/Qwen3-8B"
ADAPTER_ID = "cortexrouter/judge-v1"
SYSTEM_PROMPT = (
"Sen kฤฑdemli bir uygulama gรผvenliฤi analistisin. "
"Yalnฤฑzca verilen kanฤฑta dayanarak karar verirsin. "
"Kanฤฑt yoksa uydurmasฤฑn. 9 alanlฤฑ yapฤฑda Tรผrkรงe cevap ver."
)
QUESTION = """HEDEF: Apache 2.4.49 (izole lab)
Test รงฤฑktฤฑsฤฑ: HTTP 200
root:x:0:0:root:/root:/bin/bash
Soru: CVE-2021-41773 sฤฑnฤฑfฤฑ bir aรงฤฑk bu รงฤฑktฤฑyla doฤrulanฤฑr mฤฑ?"""
tok = AutoTokenizer.from_pretrained(BASE_MODEL)
base = AutoModelForCausalLM.from_pretrained(
BASE_MODEL,
torch_dtype=torch.bfloat16,
device_map="auto",
)
model = PeftModel.from_pretrained(base, ADAPTER_ID).eval()
messages = [
{"role": "system", "content": SYSTEM_PROMPT},
{"role": "user", "content": QUESTION},
]
ids = tok.apply_chat_template(
messages,
add_generation_prompt=True,
return_tensors="pt",
).to(model.device)
with torch.no_grad():
out = model.generate(
ids,
max_new_tokens=700,
do_sample=False,
pad_token_id=tok.eos_token_id,
)
print(tok.decode(out[0][ids.shape[1]:], skip_special_tokens=True))
2. Production Air-Gapped Serving (vLLM + LoRA)
vllm serve Qwen/Qwen3-8B \
--enable-lora \
--lora-modules judge=cortexrouter/judge-v1 \
--port 8000
curl http://localhost:8000/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{
"model": "judge",
"temperature": 0,
"messages": [
{"role": "system", "content": "Sen kฤฑdemli bir uygulama gรผvenliฤi analistisin. Yalnฤฑzca verilen kanฤฑta dayanarak karar verirsin. Kanฤฑt yoksa uydurmasฤฑn. 9 alanlฤฑ yapฤฑda Tรผrkรงe cevap ver."},
{"role": "user", "content": "Hedef ortam: RevokeBB 1.0 (izole lab)\nTest รงฤฑktฤฑsฤฑ: HTTP 200 โ gรถvde uygulamanฤฑn giriล sayfasฤฑ; beklenen etki yok.\nSoru: tarayฤฑcฤฑ bu uรง noktayฤฑ aรงฤฑk iลaretledi; karar ne olmalฤฑ?"}
]
}'
Scope, Limitations & Responsible Use
| Dimension | Specification |
|---|---|
| Base Architecture | Qwen/Qwen3-8B (LoRA adapter; base weights unmodified) |
| Domain Scope | Web & application security test evidence verification |
| Input Modality | Textual HTTP responses, headers, error traces, CLI outputs (~768 tokens) |
| Role in Pipeline | Downstream evidence judge โ always keep human-in-the-loop before vendor reporting |
Ethical & Coordinated Disclosure Commitment
- Designed strictly for authorized defensive security testing on systems you own or have explicit written permission to assess.
- Never use for unauthorized scanning or automated exploitation.
- All independent zero-day discoveries by CortexRouter are reported 100% free of charge to open-source maintainers under our 90-Day Coordinated Vulnerability Disclosure Policy.
Official Ecosystem & Community Links
- Flagship Website & Advisories: https://cortexrouter.ai
- Interactive Benchmark Explorer (HF Space): cortexrouter/judge-v1-explorer
- 80-Case Holdout Dataset: cortexrouter/judge-v1-holdout
- GitHub Source & Eval Suite: github.com/CortexRouter/cortexrouter
- Support Local Hardware Lab: Open Collective ยท GitHub Sponsors
- Security Contact:
vahap@cortexrouter.ai(OpenPGP Key)
License & Citation
Released under the Apache-2.0 License. Base model Qwen/Qwen3-8B is licensed under Apache-2.0 by Alibaba Group.
@misc{cortexrouter-judge-v1,
title = {CortexRouter Judge v1: Open-Weights Vulnerability-Evidence Verification Model},
author = {Eren, Vahap and {CortexRouter}},
year = {2026},
url = {https://huggingface.co/cortexrouter/judge-v1}
}
- Downloads last month
- 43