NeuroSploit v3.3.0 — Autonomous MD-Agent Engine

Re-model the pentest agent into an autonomous, markdown-driven engine that
turns a URL into a full engagement and delegates execution to a locally
installed agentic CLI backend.

Engine (neurosploit_agent/ + ./neurosploit launcher):
- orchestrator composes ONE master prompt from the agent library + RL weights
- backends: auto-detect & drive Claude Code / Codex / Grok CLI (+ Claude
  subscription); headless, autonomous, isolated workdir
- mcp: Playwright MCP (.mcp.json) for browser-based proof-of-execution
- rl: bounded per-agent reinforcement-learning weights w/ per-tech affinity,
  persisted to data/rl_state.json
- models: latest registry incl. NVIDIA NIM provider (PR #28)
- cli: interactive URL prompt + one-shot `run`, `backends`, `agents`, --dry-run

Agent library (agents_md/, 213 total):
- 196 vuln specialists incl. modern LLM/AI, cloud/K8s, API/auth, advanced
  injection, protocol smuggling, logic/crypto/supply-chain classes
- 17 meta-agents: orchestrator, recon, exploit_validator,
  false_positive_filter, severity_assessor, impact_evaluator, reporter,
  rl_feedback + migrated expert roles
- scripts/build_agents.py data-driven builder; REGISTRY.md index

Docs: rewritten README.md, v3.3.0 RELEASE.md, .env.example (NVIDIA NIM, xAI,
engine vars).

Retire legacy Python orchestration (neurosploit.py + agent classes) to legacy/.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

legacy/custom_agents/example_agent.py

@@ -0,0 +1,83 @@
+#!/usr/bin/env python3
+"""
+Example Custom Agent for NeuroSploitv2
+This demonstrates how to create custom agents for specific tasks
+"""
+
+import logging
+from typing import Dict
+from core.llm_manager import LLMManager
+
+logger = logging.getLogger(__name__)
+
+
+class CustomAgent:
+    """Example custom agent - Web API Security Scanner"""
+    
+    def __init__(self, config: Dict):
+        """Initialize custom agent"""
+        self.config = config
+        self.llm = LLMManager(config)
+        self.name = "WebAPIScanner"
+        logger.info(f"{self.name} initialized")
+    
+    def execute(self, target: str, context: Dict) -> Dict:
+        """Execute custom agent logic"""
+        logger.info(f"Running {self.name} on {target}")
+        
+        results = {
+            "agent": self.name,
+            "target": target,
+            "status": "running",
+            "findings": []
+        }
+        
+        try:
+            # Your custom logic here
+            # Example: API endpoint testing
+            results["findings"] = self._scan_api_endpoints(target)
+            
+            # Use AI for analysis
+            ai_analysis = self._ai_analyze(results["findings"])
+            results["ai_analysis"] = ai_analysis
+            
+            results["status"] = "completed"
+            
+        except Exception as e:
+            logger.error(f"Error in {self.name}: {e}")
+            results["status"] = "error"
+            results["error"] = str(e)
+        
+        return results
+    
+    def _scan_api_endpoints(self, target: str) -> list:
+        """Custom scanning logic"""
+        # Implement your custom scanning logic
+        return [
+            {"endpoint": "/api/users", "method": "GET", "auth": "required"},
+            {"endpoint": "/api/admin", "method": "POST", "auth": "weak"}
+        ]
+    
+    def _ai_analyze(self, findings: list) -> Dict:
+        """Use AI to analyze findings"""
+        prompt = f"""
+Analyze the following API security findings:
+
+{findings}
+
+Provide:
+1. Security assessment
+2. Risk prioritization
+3. Exploitation recommendations
+4. Remediation advice
+
+Response in JSON format.
+"""
+        
+        system_prompt = "You are an API security expert."
+        
+        try:
+            response = self.llm.generate(prompt, system_prompt)
+            return {"analysis": response}
+        except Exception as e:
+            return {"error": str(e)}