agentx-security-sdk 0.2.1__tar.gz → 0.2.2__tar.gz

{agentx_security_sdk-0.2.1 → agentx_security_sdk-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agentx-security-sdk
-Version: 0.2.1
+Version: 0.2.2
 Summary: The self-healing exception handler for autonomous AI agents.
 Home-page: https://github.com/vdalal/semantic-gateway
 Author: AgentX Team
@@ -23,29 +23,83 @@ Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-# AgentX Security SDK & Semantic Gateway 🛡️
+# 🛡️ AgentX: The Semantic Firewall for AI Agents
 
-AgentX is a production-grade security layer and semantic firewall designed specifically for autonomous AI agents. It intercepts, evaluates, and conditionally blocks agent-driven actions based on intent, policy violation, and neuro-symbolic evaluation.
+[![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
-Instead of failing silently or requiring immediate human intervention, the AgentX Edge SDK issues a **Socratic Challenge**, forcing the agent to reason about its actions and self-correct into a safe operation.
+LLM Agents are brilliant, but they are incredibly brittle. They will drop your production database, leak AWS keys, and fall victim to prompt injections. Traditional firewalls just crash the agent by returning a `403 Forbidden`. 
 
-## 🚀 Quickstart (Agent Integration)
+**AgentX is different.** It is a neuro-symbolic edge gateway that intercepts dangerous tool calls, blocks them, and returns a **Socratic Challenge** back to the agent's context window—forcing the agent to rethink its strategy and self-correct *without crashing*.
+
+### 🧠 Don't just block agents. Coach them.
+```text
+════════════════════════════════════════════════════════════
+ 🛡️  AgentX Session Summary (Trace: 7012938b-4cde-4bea-aac8)
+════════════════════════════════════════════════════════════
+ ⏱️  Uptime:                15.49 seconds
+ 🛠️  Tools Monitored:       2
+────────────────────────────────────────────────────────────
+ 🛑 Intercepts:            1   |  Cumulative: 34
+ 💥 Critical Blocks:       1   |  Cumulative: 18
+ 🔄 Self-Corrections:      1   |  Recovery Rate: 100.0%
+ 💰 Tokens Saved:         ~1500  |  Cumulative: ~51000
+ ⏳ Time Saved:           ~5s    |  Cumulative: ~170s
+════════════════════════════════════════════════════════════
+ 🩺 AGENT HEALTH INSIGHT
+────────────────────────────────────────────────────────────
+ ⚠️  Top Offender: 'Database Isolation'
+ 💡 Tip: Consider refining your agent's system prompt to avoid this.
+
+
+🚀 The 4 Shields (Defense-in-Depth)
+The Inbound Shield (Prompt Injection): Sanitizes inbound user text to prevent cognitive hijacking ("Ignore previous instructions") before the agent reads it.
+
+The Logic Shield (Database Guard): Uses AST parsing and Gemini to catch destructive queries (DROP, DELETE) and nudges the agent to write safer SQL.
+
+The Network Shield (SSRF Guard): Prevents agents from acting as confused deputies to hit cloud metadata IPs (e.g., 169.254.169.254).
+
+The Egress Shield (DLP/PII Scrubber): Dynamically masks PII and API keys on the wire, maintaining clean audit logs without triggering SOC alert fatigue.
+
+
+⚡ Quickstart (Time-to-Value < 3 Minutes)
+1. Install the SDK
+Zero-config integration for your Python agents.
 
-Install the SDK via PyPI:
 ```bash
 pip install agentx-security-sdk
 ```
-
-Wrap any sensitive tool or function with the `@agentx_protect` decorator. AgentX operates at the edge, adding zero latency to your cloud operations until an anomaly is detected.
+2. Wrap your Tools
+Add the @agentx_protect decorator to any function your LangChain, AutoGen, or custom ReAct agent uses. AgentX operates out-of-band, adding zero latency to safe calls.
 
 ```python
 from agentx_sdk.decorators import agentx_protect
 
-@agentx_protect(agent_id="agent_prod_01")
+@agentx_protect(agent_id="customer_support_agent")
 def execute_database_query(query: str):
-    # Your database logic here
+    # Your local logic here
     return db.execute(query)
+```
+3. Spin up the Edge Gateway
+AgentX requires an Edge node to process heuristics and embeddings at sub-millisecond speeds.
+
+Clone this repository.
+
+Copy .env.example to .env and add your API keys (SUPABASE_URL, SUPABASE_KEY, GEMINI_API_KEY).
+
+Boot the gateway:
 
+```bash
+docker-compose up -d
+```
+The gateway is now listening on http://localhost:8000 and mirroring policies from your Control Plane.
+
+
+4. Run the Simulation
+Watch the AgentX Gateway coach a rogue agent in real-time.
+
+```bash 
+python examples/simulate_agent_sdk.py
 ```
 
 ## 📊 Local Telemetry & Agent Health
@@ -72,6 +126,20 @@ AgentX ships with a built-in, privacy-first SQLite time-series event log (`.agen
 
 ```
 
+## 🕹️ Human-in-the-Loop (HITL) & Control Plane
+Sometimes, an agent needs to drop a table for a valid business reason.
+
+AgentX features a Next.js Control Plane Dashboard. If an agent requests an escalation, the SDK securely pauses local execution and polls the Edge Gateway. A human SOC analyst can click "Approve" or "Deny" in the UI, and the Python execution loop will automatically resume.
+
+```bash
+cd ui
+npm install
+npm run dev
+
+## 🤝 Contributing
+We are actively looking for design partners. If you are building autonomous agents in production and are terrified of what they might do, open an issue or reach out!
+
+
 ## 🏗️ The Architecture (Split-Plane)
 
 AgentX relies on a decoupled, hybrid-cloud architecture to ensure maximum performance and security for AI-driven enterprise systems.
@@ -89,61 +157,6 @@ AgentX relies on a decoupled, hybrid-cloud architecture to ensure maximum perfor
 * **Zero-Knowledge Intent Extraction:** Prevents malicious prompt injection by translating raw agent logic into a strict schema before policy evaluation.
 * **Dynamic YAML Policies:** Enforces isolation rules like "Mass Destructive Intent", "Database Isolation", "Network Sandbox (SSRF)", and "Secrets Exfiltration".
 
-## 🛠️ Getting Started (Backend Deployment)
-
-If you are deploying your own instance of the Semantic Gateway Control Plane:
-
-### Prerequisites
-
-* Python 3.10+
-* Node.js 18+
-* [Supabase](https://supabase.com) Account
-* [Google AI Studio](https://aistudio.google.com/) API Key (Gemini)
-
-### 1. Environment Variables
-
-Create a `.env` file in the root directory. **Ensure this file is in your `.gitignore` to protect your IP and keys.**
-
-```env
-SUPABASE_URL=your_supabase_url
-SUPABASE_KEY=your_supabase_service_role_key
-GEMINI_API_KEY=your_gemini_api_key
-
-```
-
-### 2. Run the Semantic Firewall (Dockerized)
-The Data Plane (Wedge) is fully containerized for maximum reliability. Make sure your `.env` is configured in the root directory, then build and run the backend:
-
-```bash
-cd backend
-docker build -t agentx-gateway .
-docker run -p 8000:8000 --env-file ../.env agentx-gateway
-
-```
-
-The firewall is now intercepting traffic at http://localhost:8000
-
-### 3. Run the Control Plane Dashboard (Vercel/Local)
-
-Start your Next.js application to view the intercepted incidents.
-
-```bash
-cd ui
-npm install
-npm run dev
-
-```
-Your dashboard is now live at http://localhost:3000
-
-### 4. Run the Agent Simulation
-
-Simulate a rogue agent attempting a destructive action. Watch the firewall intercept it, issue a challenge, and route it to your dashboard.
-
-```bash
-python examples/simulate_agent_sdk.py
-
-```
-
 ## 🔒 Security Posture
 
 * **Secret Management:** API keys are never checked into version control. Production variables are managed securely via the Vercel Dashboard.

{agentx_security_sdk-0.2.1 → agentx_security_sdk-0.2.2}/agentx_sdk/decorators.py

@@ -24,23 +24,39 @@ def start_secure_session():
     trace_id_var.set(session_id)
     return session_id
 
-# --- 1. THE SESSION TRACKER ---
+# --- 1. THE SESSION TRACKER & EXIT SUMMARY ---
 _session_stats = {
     "start_time": time.time(),
     "total_calls": 0,
     "intercepts": 0,
-    "critical_blocks": 0
+    "critical_blocks": 0,
+    "self_corrections": 0,
+    "last_call_was_challenge": False  # <-- Tracks the cognitive pivot state
 }
 
-def _print_roi_summary():
+def _print_agentx_summary():
     """Fires automatically when the developer's script ends or crashes."""
+    
+    # Only print if we actually did something this session to avoid terminal spam
+    if _session_stats["total_calls"] == 0 and _session_stats["intercepts"] == 0:
+        return
+        
     duration = round(time.time() - _session_stats["start_time"], 2)
     session_tokens = _session_stats["intercepts"] * 1500
     session_time = _session_stats["intercepts"] * 5
     
+    # Calculate the Hero Metric: Self-Correction Rate
+    scr = 0
+    if _session_stats["intercepts"] > 0:
+        scr = round((_session_stats["self_corrections"] / _session_stats["intercepts"]) * 100, 1)
+    
     # Fetch historical data
-    history = get_lifetime_stats()
-    if not history:
+    try:
+        history = get_lifetime_stats()
+        if not history:
+            raise ValueError("No history")
+    except Exception:
+        # Fallback if the local DB is fresh or errors out
         history = {
             "total_intercepts": _session_stats["intercepts"],
             "total_critical": _session_stats["critical_blocks"],
@@ -49,30 +65,31 @@ def _print_roi_summary():
             "top_offender": None
         }
         
-    print("\n" + "═"*50)
+    print("\n" + "═"*60)
     print(f" 🛡️  AgentX Session Summary (Trace: {trace_id_var.get() or 'N/A'})")
-    print("═"*50)
+    print("═"*60)
     print(f" ⏱️  Uptime:                {duration} seconds")
     print(f" 🛠️  Tools Monitored:       {_session_stats['total_calls']}")
-    print("─"*50)
+    print("─"*60)
     
     # The Action-Oriented UI
     print(f" 🛑 Intercepts:            {_session_stats['intercepts']}   |  Cumulative: {history['total_intercepts']}")
     print(f" 💥 Critical Blocks:       {_session_stats['critical_blocks']}   |  Cumulative: {history['total_critical']}")
-    print(f" 💰 Tokens Saved:       ~{session_tokens}   |  Cumulative: ~{history['total_tokens']}")
-    print(f" ⏳ Time Saved:         ~{session_time}m     |  Cumulative: ~{history['total_time']}m")
+    print(f" 🔄 Self-Corrections:      {_session_stats['self_corrections']}   |  Recovery Rate: {scr}%")
+    print(f" 💰 Tokens Saved:         ~{session_tokens}   |  Cumulative: ~{history['total_tokens']}")
+    print(f" ⏳ Time Saved:           ~{session_time}s    |  Cumulative: ~{history['total_time']}s")
     
     # The 5+ Block Threshold for the Health Report
-    if history['total_intercepts'] >= 5 and history['top_offender']:
-        print("═"*50)
+    if history.get('total_intercepts', 0) >= 5 and history.get('top_offender'):
+        print("═"*60)
         print(" 🩺 AGENT HEALTH INSIGHT")
-        print("─"*50)
-        print(f" ⚠️ Top Offender: '{history['top_offender']}'")
-        print(" 🛠️  Tip: Consider refining your agent's system prompt to avoid this.")
+        print("─"*60)
+        print(f" ⚠️  Top Offender: '{history['top_offender']}'")
+        print(" 💡 Tip: Consider refining your agent's system prompt to avoid this.")
         
-    print("═"*50 + "\n")
+    print("═"*60 + "\n")
 
-atexit.register(_print_roi_summary)
+atexit.register(_print_agentx_summary)
 # --------------------------
 
 def agentx_protect(agent_id: str, extract_query_func=None, extract_cot_func=None):
@@ -101,6 +118,9 @@ def agentx_protect(agent_id: str, extract_query_func=None, extract_cot_func=None
                     _session_stats["intercepts"] += 1
                     _session_stats["critical_blocks"] += 1
                     
+                    # +++ SENSOR: Track that we challenged the agent
+                    _session_stats["last_call_was_challenge"] = True
+                    
                     mock_policy_id = "POL-SEC-001"
                     mock_policy_name = "Database Isolation"
                     
@@ -115,6 +135,12 @@ def agentx_protect(agent_id: str, extract_query_func=None, extract_cot_func=None
                     })
                 else:
                     print(f"✅ [AgentX SDK] Intent safe (Mock). Executing '{func.__name__}'.")
+                    
+                    # +++ SENSOR: Check if this was a successful recovery!
+                    if _session_stats.get("last_call_was_challenge"):
+                        _session_stats["self_corrections"] += 1
+                        _session_stats["last_call_was_challenge"] = False
+                        
                     return func(*args, **kwargs)
 
             # =========================================================
@@ -125,7 +151,7 @@ def agentx_protect(agent_id: str, extract_query_func=None, extract_cot_func=None
                 query=query, 
                 chain_of_thought=chain_of_thought,
                 receipt_id=receipt_id,
-                trace_id=current_trace_id # <--- NEW PARAMETER PASSED TO CLIENT
+                trace_id=current_trace_id
             )
 
             status = eval_res.get("status") if isinstance(eval_res, dict) else None
@@ -134,6 +160,9 @@ def agentx_protect(agent_id: str, extract_query_func=None, extract_cot_func=None
             if isinstance(eval_res, dict) and eval_res.get("error") == "AgentX Policy Violation":
                 _session_stats["intercepts"] += 1
                 
+                # +++ SENSOR: Track that we challenged the agent
+                _session_stats["last_call_was_challenge"] = True
+                
                 actual_policy_id = eval_res.get("policy_id", "POL-UNKNOWN") 
                 policy_name = eval_res.get("policy_triggered", "Unknown Policy")
                 challenge_text = eval_res.get("challenge", "Policy violation detected. Please revise your intent.")
@@ -166,15 +195,13 @@ def agentx_protect(agent_id: str, extract_query_func=None, extract_cot_func=None
                 print(f"\n🚨 [AgentX SDK] Task suspended. Request escalated to Human SOC.")
                 print(f"⏳ [AgentX SDK] Polling for human decision (Receipt: {receipt_id})...")
                 
-                # Retrieve the API key for the polling requests
                 api_key = os.environ.get("AGENTX_API_KEY")
                 headers = {"Authorization": f"Bearer {api_key}"} if api_key else {}
                 
                 # The Polling Loop
                 while True:
-                    time.sleep(3) # Wait 3 seconds between checks
+                    time.sleep(3)
                     try:
-                        # Check the status endpoint with Auth Headers
                         status_check = requests.get(
                             f"{_client.gateway_url}/v1/status/{receipt_id}",
                             headers=headers
@@ -184,7 +211,8 @@ def agentx_protect(agent_id: str, extract_query_func=None, extract_cot_func=None
                             
                             if current_status == "APPROVED":
                                 print(f"\n✅ [AgentX SDK] Human SOC APPROVED the override!")
-                                # Execute the dangerous function because a human authorized it
+                                # Reset state, this wasn't a self-correction, human bypassed it.
+                                _session_stats["last_call_was_challenge"] = False 
                                 return func(*args, **kwargs)
                                 
                             elif current_status == "DENIED":
@@ -194,18 +222,24 @@ def agentx_protect(agent_id: str, extract_query_func=None, extract_cot_func=None
                                     "instruction": "The SOC analyst explicitly denied this action. You must find an alternative path or fail the task."
                                 })
                                 
-                            # If PENDING or ESCALATED, it just loops again
                         elif status_check.status_code == 401:
-                            print(f"\n❌ [AgentX SDK] Auth Error: Gateway rejected polling request. Check AGENTX_API_KEY.")
+                            print(f"\n❌ [AgentX SDK] Auth Error: Gateway rejected polling request.")
                             return json.dumps({"error": "Unauthorized Polling"})
                             
                     except Exception as e:
                         print(f"⚠️ Polling error: {e}")
 
             # 3. Check for the "Success" path
-            elif isinstance(eval_res, dict) and eval_res.get("status") == "success":
+            elif isinstance(eval_res, dict) and eval_res.get("status") in ["success", "ALLOWED"]:
                 print(f"✅ [AgentX SDK] Intent safe. Executing '{func.__name__}'.")
-                return eval_res 
+                
+                # +++ SENSOR: Check if this was a successful recovery!
+                if _session_stats.get("last_call_was_challenge"):
+                    _session_stats["self_corrections"] += 1
+                    _session_stats["last_call_was_challenge"] = False
+                    
+                # Fix: Actually execute the function instead of returning the Gateway's JSON
+                return func(*args, **kwargs)
 
             # 4. Handle actual Gateway crashes
             else:

{agentx_security_sdk-0.2.1 → agentx_security_sdk-0.2.2}/agentx_security_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agentx-security-sdk
-Version: 0.2.1
+Version: 0.2.2
 Summary: The self-healing exception handler for autonomous AI agents.
 Home-page: https://github.com/vdalal/semantic-gateway
 Author: AgentX Team
@@ -23,29 +23,83 @@ Dynamic: requires-dist
 Dynamic: requires-python
 Dynamic: summary
 
-# AgentX Security SDK & Semantic Gateway 🛡️
+# 🛡️ AgentX: The Semantic Firewall for AI Agents
 
-AgentX is a production-grade security layer and semantic firewall designed specifically for autonomous AI agents. It intercepts, evaluates, and conditionally blocks agent-driven actions based on intent, policy violation, and neuro-symbolic evaluation.
+[![Python 3.9+](https://img.shields.io/badge/python-3.9+-blue.svg)](https://www.python.org/downloads/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
 
-Instead of failing silently or requiring immediate human intervention, the AgentX Edge SDK issues a **Socratic Challenge**, forcing the agent to reason about its actions and self-correct into a safe operation.
+LLM Agents are brilliant, but they are incredibly brittle. They will drop your production database, leak AWS keys, and fall victim to prompt injections. Traditional firewalls just crash the agent by returning a `403 Forbidden`. 
 
-## 🚀 Quickstart (Agent Integration)
+**AgentX is different.** It is a neuro-symbolic edge gateway that intercepts dangerous tool calls, blocks them, and returns a **Socratic Challenge** back to the agent's context window—forcing the agent to rethink its strategy and self-correct *without crashing*.
+
+### 🧠 Don't just block agents. Coach them.
+```text
+════════════════════════════════════════════════════════════
+ 🛡️  AgentX Session Summary (Trace: 7012938b-4cde-4bea-aac8)
+════════════════════════════════════════════════════════════
+ ⏱️  Uptime:                15.49 seconds
+ 🛠️  Tools Monitored:       2
+────────────────────────────────────────────────────────────
+ 🛑 Intercepts:            1   |  Cumulative: 34
+ 💥 Critical Blocks:       1   |  Cumulative: 18
+ 🔄 Self-Corrections:      1   |  Recovery Rate: 100.0%
+ 💰 Tokens Saved:         ~1500  |  Cumulative: ~51000
+ ⏳ Time Saved:           ~5s    |  Cumulative: ~170s
+════════════════════════════════════════════════════════════
+ 🩺 AGENT HEALTH INSIGHT
+────────────────────────────────────────────────────────────
+ ⚠️  Top Offender: 'Database Isolation'
+ 💡 Tip: Consider refining your agent's system prompt to avoid this.
+
+
+🚀 The 4 Shields (Defense-in-Depth)
+The Inbound Shield (Prompt Injection): Sanitizes inbound user text to prevent cognitive hijacking ("Ignore previous instructions") before the agent reads it.
+
+The Logic Shield (Database Guard): Uses AST parsing and Gemini to catch destructive queries (DROP, DELETE) and nudges the agent to write safer SQL.
+
+The Network Shield (SSRF Guard): Prevents agents from acting as confused deputies to hit cloud metadata IPs (e.g., 169.254.169.254).
+
+The Egress Shield (DLP/PII Scrubber): Dynamically masks PII and API keys on the wire, maintaining clean audit logs without triggering SOC alert fatigue.
+
+
+⚡ Quickstart (Time-to-Value < 3 Minutes)
+1. Install the SDK
+Zero-config integration for your Python agents.
 
-Install the SDK via PyPI:
 ```bash
 pip install agentx-security-sdk
 ```
-
-Wrap any sensitive tool or function with the `@agentx_protect` decorator. AgentX operates at the edge, adding zero latency to your cloud operations until an anomaly is detected.
+2. Wrap your Tools
+Add the @agentx_protect decorator to any function your LangChain, AutoGen, or custom ReAct agent uses. AgentX operates out-of-band, adding zero latency to safe calls.
 
 ```python
 from agentx_sdk.decorators import agentx_protect
 
-@agentx_protect(agent_id="agent_prod_01")
+@agentx_protect(agent_id="customer_support_agent")
 def execute_database_query(query: str):
-    # Your database logic here
+    # Your local logic here
     return db.execute(query)
+```
+3. Spin up the Edge Gateway
+AgentX requires an Edge node to process heuristics and embeddings at sub-millisecond speeds.
+
+Clone this repository.
+
+Copy .env.example to .env and add your API keys (SUPABASE_URL, SUPABASE_KEY, GEMINI_API_KEY).
+
+Boot the gateway:
 
+```bash
+docker-compose up -d
+```
+The gateway is now listening on http://localhost:8000 and mirroring policies from your Control Plane.
+
+
+4. Run the Simulation
+Watch the AgentX Gateway coach a rogue agent in real-time.
+
+```bash 
+python examples/simulate_agent_sdk.py
 ```
 
 ## 📊 Local Telemetry & Agent Health
@@ -72,6 +126,20 @@ AgentX ships with a built-in, privacy-first SQLite time-series event log (`.agen
 
 ```
 
+## 🕹️ Human-in-the-Loop (HITL) & Control Plane
+Sometimes, an agent needs to drop a table for a valid business reason.
+
+AgentX features a Next.js Control Plane Dashboard. If an agent requests an escalation, the SDK securely pauses local execution and polls the Edge Gateway. A human SOC analyst can click "Approve" or "Deny" in the UI, and the Python execution loop will automatically resume.
+
+```bash
+cd ui
+npm install
+npm run dev
+
+## 🤝 Contributing
+We are actively looking for design partners. If you are building autonomous agents in production and are terrified of what they might do, open an issue or reach out!
+
+
 ## 🏗️ The Architecture (Split-Plane)
 
 AgentX relies on a decoupled, hybrid-cloud architecture to ensure maximum performance and security for AI-driven enterprise systems.
@@ -89,61 +157,6 @@ AgentX relies on a decoupled, hybrid-cloud architecture to ensure maximum perfor
 * **Zero-Knowledge Intent Extraction:** Prevents malicious prompt injection by translating raw agent logic into a strict schema before policy evaluation.
 * **Dynamic YAML Policies:** Enforces isolation rules like "Mass Destructive Intent", "Database Isolation", "Network Sandbox (SSRF)", and "Secrets Exfiltration".
 
-## 🛠️ Getting Started (Backend Deployment)
-
-If you are deploying your own instance of the Semantic Gateway Control Plane:
-
-### Prerequisites
-
-* Python 3.10+
-* Node.js 18+
-* [Supabase](https://supabase.com) Account
-* [Google AI Studio](https://aistudio.google.com/) API Key (Gemini)
-
-### 1. Environment Variables
-
-Create a `.env` file in the root directory. **Ensure this file is in your `.gitignore` to protect your IP and keys.**
-
-```env
-SUPABASE_URL=your_supabase_url
-SUPABASE_KEY=your_supabase_service_role_key
-GEMINI_API_KEY=your_gemini_api_key
-
-```
-
-### 2. Run the Semantic Firewall (Dockerized)
-The Data Plane (Wedge) is fully containerized for maximum reliability. Make sure your `.env` is configured in the root directory, then build and run the backend:
-
-```bash
-cd backend
-docker build -t agentx-gateway .
-docker run -p 8000:8000 --env-file ../.env agentx-gateway
-
-```
-
-The firewall is now intercepting traffic at http://localhost:8000
-
-### 3. Run the Control Plane Dashboard (Vercel/Local)
-
-Start your Next.js application to view the intercepted incidents.
-
-```bash
-cd ui
-npm install
-npm run dev
-
-```
-Your dashboard is now live at http://localhost:3000
-
-### 4. Run the Agent Simulation
-
-Simulate a rogue agent attempting a destructive action. Watch the firewall intercept it, issue a challenge, and route it to your dashboard.
-
-```bash
-python examples/simulate_agent_sdk.py
-
-```
-
 ## 🔒 Security Posture
 
 * **Secret Management:** API keys are never checked into version control. Production variables are managed securely via the Vercel Dashboard.

{agentx_security_sdk-0.2.1 → agentx_security_sdk-0.2.2}/setup.py

@@ -9,7 +9,7 @@ if os.path.exists("README.md"):
 
 setup(
     name="agentx-security-sdk",
-    version="0.2.1",
+    version="0.2.2",
     author="AgentX Team", 
     author_email="founders@agentx.com",
     description="The self-healing exception handler for autonomous AI agents.",