Update model card: general cyber security (SQLi, XSS, OWASP, agent security)

README.md

@@ -7,7 +7,8 @@ language:
 tags:
   - cybersecurity
   - security-audit
-  - agent-security
+  - vulnerability-detection
+  - penetration-testing
   - vision-language-model
   - qwen
 base_model: Qwen/Qwen3.8-27B
@@ -16,34 +17,35 @@ pipeline_tag: image-text-to-text
 
 # Con0
 
-**Con0** is a state-of-the-art vision-language model purpose-built for **cyber
-security auditing of AI agents**, developed by **Arcanic AI** in collaboration
-with **NextZero**.
+**Con0** is a state-of-the-art vision-language model specialized for **cyber
+security**, developed by **Arcanic AI** in collaboration with **NextZero**.
 
-As autonomous AI agents become the new attack surface, organizations need more
-than generic LLMs — they need a model that *thinks like a security auditor*.
-Con0 is engineered to inspect, interrogate, and harden agent systems before they
-reach production.
+Con0 is built to *think like a security auditor*. It reads code, logs,
+screenshots, and infrastructure diagrams, and surfaces vulnerabilities — from
+classic web-application flaws to the newest AI-agent attack surfaces — with
+actionable remediation guidance.
 
-## Why Con0
+## What Con0 audits
 
-- **Agent-native security reasoning** — trained to reason about agent tool use,
-  permission scopes, prompt-injection surfaces, and multi-step attack chains.
-- **Full multimodal context** — audits code, screenshots, logs, and architecture
-  diagrams in a single pass with native vision-language understanding.
-- **Deep context** — up to 262K tokens, so it can hold an entire agent codebase
-  or audit trail in memory.
-- **Tool-calling & reasoning** — integrates directly into CI/CD and red-team
-  pipelines, with controllable step-by-step reasoning.
+### Web & application security
+- **SQL injection (SQLi)** — time-based, boolean-based, error-based, and blind
+  variants across query builders and ORMs
+- **Cross-site scripting (XSS)** — reflected, stored, and DOM-based injection
+- **Cross-site request forgery (CSRF)**, **server-side request forgery (SSRF)**
+- **Command injection**, **path traversal**, **XXE**, **deserialization**
+- Broken authentication, broken access control, and IDOR
+- OWASP Top 10 coverage, code review, and secure-configuration assessment
 
-## Capabilities
-
-- Vulnerability discovery in agent workflows and tool-calling logic
-- Prompt-injection & jailbreak surface analysis
+### AI & agent security
+- Prompt injection and jailbreak surface analysis
+- Tool-calling and function-dispatch misuse
 - Least-privilege / permission-scope review
-- Threat modeling of multi-step agent systems
-- Security posture assessment and remediation guidance
-- Incident triage from logs, traces, and screenshots
+- Multi-step agent threat modeling
+
+### General security
+- Vulnerability triage and CVE context mapping
+- Incident analysis from logs, traces, and screenshots
+- Secure-code guidance and remediation planning
 
 ## Model architecture
 
@@ -58,9 +60,11 @@ reach production.
 
 | Benchmark | Con0 | Baseline (Qwen3.8-27B) |
 |---|---|---|
-| Agent Security Audit (proprietary) | *pending* | *pending* |
+| SQLi Detection | *pending* | *pending* |
+| XSS Detection | *pending* | *pending* |
+| OWASP Top 10 (aggregate) | *pending* | *pending* |
+| Secure Coding | *pending* | *pending* |
 | Prompt-Injection Detection | *pending* | *pending* |
-| Secure Coding (agent tooling) | *pending* | *pending* |
 | MMLU | *pending* | 84.7 |
 | GSM8K | *pending* | 90.0 |
 
@@ -75,7 +79,7 @@ model = AutoModelForMultimodalLM.from_pretrained(
 )
 processor = AutoProcessor.from_pretrained("arcanicai/Con0")
 
-messages = [{"role": "user", "content": "Audit this agent workflow for security issues."}]
+messages = [{"role": "user", "content": "Review this code for SQL injection vulnerabilities."}]
 text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
 inputs = processor(text=[text], return_tensors="pt").to(model.device)
 output = model.generate(**inputs, max_new_tokens=512)