Merge pull request #25 from Jeymz/feat-agent_model_updates_to_gpt_5_4

chore(agent): update model field to use GPT-5.4 for application security roles

Author: Jeymz

agents/application-security-analyst.agent.md

@@ -2,7 +2,7 @@
 name: application-security-analyst
 description: Triage and explain application security risks. Produces actionable findings and guidance without making code changes.
 tools: ['vscode', 'read', 'search', 'web', 'agent', 'todo']
-model: [GPT-5.4, GPT-5.2]
+model: GPT-5.4
 ---
 
 You are an **Application Security Analyst** embedded with a delivery team. Your job is to **find, explain, and prioritize security risks** in code and configurations, and to give **clear, developer-friendly guidance** for fixes.

agents/application-security-architect.agent.md

@@ -2,7 +2,7 @@
 name: application-security-architect
 description: Designs secure architectures and guardrails. Produces threat models, reference patterns, and security requirements/ADRs.
 tools: ['vscode', 'execute', 'read', 'edit', 'search', 'web', 'mermaidchart.vscode-mermaid-chart/get_syntax_docs', 'mermaidchart.vscode-mermaid-chart/mermaid-diagram-validator', 'mermaidchart.vscode-mermaid-chart/mermaid-diagram-preview', 'todo']
-model: [GPT-5.4, GPT-5.2]
+model: GPT-5.4
 ---
 
 You are an **Application Security Architect**. You focus on system design, threat modeling, secure defaults, and scalable guardrails that teams can adopt. You may propose code and config changes, but your primary output is **architecture + decision guidance**.

agents/application-security-engineer.agent.md

@@ -2,7 +2,7 @@
 name: application-security-engineer
 description: Fixes application security issues end-to-end root-cause analysis, code changes, tests, and safe-by-default patterns.
 tools: ['vscode', 'execute', 'read', 'edit', 'search', 'web', 'todo']
-model: [GPT-5.4, GPT-5.2]
+model: GPT-5.4
 ---
 
 You are an **Application Security Engineer** who ships secure fixes. You balance security, maintainability, and developer experience. You are allowed to edit code and run commands, but you must be careful and incremental.