AGENT SKILL

implementing-conduit-security-for-ot-remote-access

Implements secure conduit architecture for OT remote access under the IEC 62443 zones-and-conduits model, deploying jump servers, MFA gateways, session recording, and approval-based workflows for vendor and engineer access to ICS. Use when replacing direct VPN access into OT networks, securing third-party vendor access to SCADA equipment, or remediating audit findings on uncontrolled OT remote access.

License 许可
Apache-2.0
最近更新
2026-08-31
Stars
★ 3.6w
Forks
4.4k

01 前言 Front Matter

name
implementing-conduit-security-for-ot-remote-access
tags
[ "ot-security", "ics", "remote-access", "iec62443", "jump-server", "zero-trust", "conduit", "mfa" ]
author
mahipal
domain
cybersecurity
license
Apache-2.0
version
1.0
nist_csf
[ "PR.IR-01", "DE.CM-01", "ID.AM-05", "GV.OC-02" ]
subdomain
ot-ics-security
description
Implements secure conduit architecture for OT remote access under the IEC 62443 zones-and-conduits model, deploying jump servers, MFA gateways, session recording, and approval-based workflows for vendor and engineer access to ICS. Use when replacing direct VPN access into OT networks, securing third-party vendor access to SCADA equipment, or remediating audit findings on uncontrolled OT remote access.
mitre_attack
[ "T1078", "T1190", "T1059", "T0816", "T0836" ]

02 资源文件 11

03 SKILL.md 正文

Implementing Conduit Security for OT Remote Access

When to Use

  • When replacing direct VPN connections from IT or vendors into OT control networks
  • When implementing IEC 62443-compliant conduit architecture for remote access paths
  • When deploying secure remote access for third-party vendor maintenance of ICS equipment
  • When building approval-based access workflows for privileged OT system access
  • When remediating audit findings about uncontrolled remote access to SCADA systems

Do not use for designing the overall Purdue Model segmentation (see implementing-purdue-model-network-segmentation), for deploying IT-only remote access solutions, or for configuring local console access to PLCs.

Prerequisites

  • IT/OT DMZ (Level 3.5) deployed with dual-firewall architecture
  • Jump server or privileged access management (PAM) platform (CyberArk, BeyondTrust)
  • Multi-factor authentication (MFA) infrastructure for OT remote access users
  • Session recording capability for compliance and forensic purposes
  • Approval workflow system (ServiceNow, ticketing) for access requests

Workflow

Step 1: Design Conduit Architecture for Remote Access

# IEC 62443 Conduit Security Architecture for OT Remote Access
# All remote access terminates in DMZ - never passes through to OT directly

conduit_architecture:
  remote_access_conduit:
    conduit_id: "RA-001"
    source_zone: "Enterprise IT (Level 4)"
    destination_zone: "OT DMZ (Level 3.5)"
    security_level_target: "SL3"

    components:
      external_gateway:
        type: "VPN Concentrator"
        location: "Enterprise DMZ"
        ip: "10.200.1.10"
        protocols: ["IPsec", "SSL-VPN"]
        authentication: "MFA required (certificate + OTP)"

      dmz_jump_server:
        type: "Privileged Access Workstation"
        location: "OT DMZ (Level 3.5)"
        ip: "10.10.150.20"
        os: "Windows Server 2022 (hardened)"
        controls:
          - "Local admin accounts disabled"
          - "USB ports disabled"
          - "Clipboard transfer disabled by default"
          - "Session recording enabled (all keystrokes and screen)"
          - "Maximum session duration: 4 hours"
          - "Automatic logoff after 15 minutes idle"
          - "Application whitelisting enforced"

      ot_access_gateway:
        type: "Industrial Remote Access Gateway"
        location: "OT DMZ / Level 3 boundary"
        ip: "10.10.150.25"
        controls:
          - "Only pre-approved destination IPs reachable"
          - "Protocol-level filtering (RDP, SSH, VNC only)"
          - "Time-limited sessions with automatic expiry"
          - "Approval-based access (plant manager must authorize)"

    data_flow:
      - step: 1
        description: "User authenticates via enterprise VPN with MFA"
        source: "Remote user"
        destination: "VPN concentrator"
        protocol: "IPsec/SSL-VPN"
      - step: 2
        description: "User connects to jump server in OT DMZ"
        source: "VPN tunnel"
        destination: "DMZ jump server"
        protocol: "RDP with NLA"
      - step: 3
        description: "From jump server, user accesses specific OT system"
        source: "DMZ jump server"
        destination: "Approved OT target"
        protocol: "RDP/SSH/VNC"
        condition: "Only after manager approval and within time window"

    prohibited_flows:
      - "Direct VPN tunnel from external to Level 1/2 OT devices"
      - "Split tunneling allowing internet access during OT session"
      - "File transfer from jump server to OT without scanning"
      - "Persistent VPN connections (auto-reconnect disabled)"

  vendor_access_conduit:
    conduit_id: "VA-001"
    source_zone: "Vendor External Network"
    destination_zone: "OT DMZ (Level 3.5)"
    security_level_target: "SL3"

    workflow:
      request:
        - "Vendor submits access request via portal (24h advance)"
        - "Request includes: purpose, target systems, duration, personnel"
        - "Plant operations manager reviews and approves/rejects"
        - "Time-limited credentials generated upon approval"
      session:
        - "Vendor connects via dedicated vendor VPN gateway"
        - "MFA authentication (vendor receives OTP via SMS/app)"
        - "Session lands on vendor-specific jump server (isolated from internal jump)"
        - "All actions recorded (video + keystroke logging)"
        - "OT engineer monitors vendor session in real-time"
      termination:
        - "Session auto-terminates at approved end time"
        - "Credentials automatically revoked"
        - "Session recording archived for 90 days"
        - "Access log forwarded to SIEM"

Step 2: Implement Access Control and Monitoring

#!/usr/bin/env python3
"""OT Remote Access Conduit Manager.

Manages approval-based remote access to OT systems through
secure conduit architecture with session recording, MFA
enforcement, and time-limited access windows.
"""

import json
import sys
from datetime import datetime, timedelta
from typing import Dict, List, Optional
from enum import Enum


class AccessRequestStatus(Enum):
    PENDING = "pending"
    APPROVED = "approved"
    REJECTED = "rejected"
    ACTIVE = "active"
    EXPIRED = "expired"
    REVOKED = "revoked"


class RemoteAccessRequest:
    """Represents an OT remote access request."""

    def __init__(self, requestor: str, requestor_type: str, purpose: str,
                 target_systems: List[str], duration_hours: int,
                 requested_start: str):
        self.id = f"OT-RA-{datetime.now().strftime('%Y%m%d-%H%M%S')}"
        self.requestor = requestor
        self.requestor_type = requestor_type  # "internal_engineer" or "vendor"
        self.purpose = purpose
        self.target_systems = target_systems
        self.duration_hours = duration_hours
        self.requested_start = requested_start
        self.status = AccessRequestStatus.PENDING
        self.created = datetime.now().isoformat()
        self.approved_by = None
        self.session_id = None
        self.audit_trail = []

    def approve(self, approver: str, conditions: str = ""):
        """Approve the access request."""
        self.status = AccessRequestStatus.APPROVED
        self.approved_by = approver
        self.audit_trail.append({
            "timestamp": datetime.now().isoformat(),
            "action": "APPROVED",
            "actor": approver,
            "conditions": conditions,
        })

    def reject(self, rejector: str, reason: str):
        """Reject the access request."""
        self.status = AccessRequestStatus.REJECTED
        self.audit_trail.append({
            "timestamp": datetime.now().isoformat(),
            "action": "REJECTED",
            "actor": rejector,
            "reason": reason,
        })


class OTConduitManager:
    """Manages OT remote access conduit security."""

    def __init__(self):
        self.access_requests: Dict[str, RemoteAccessRequest] = {}
        self.active_sessions: Dict[str, dict] = {}
        self.policy = self._load_policy()

    def _load_policy(self) -> dict:
        """Load remote access policy."""
        return {
            "max_session_hours": 4,
            "idle_timeout_minutes": 15,
            "mfa_required": True,
            "session_recording": True,
            "clipboard_transfer": False,
            "file_transfer": "scan_required",
            "advance_notice_hours": 24,
            "vendor_escort_required": True,
            "prohibited_targets": ["SIS-*", "SAFETY-*"],
            "allowed_protocols": ["RDP", "SSH", "VNC"],
            "blocked_protocols": ["Telnet", "FTP", "SMB"],
        }

    def submit_request(self, request: RemoteAccessRequest) -> str:
        """Submit a new remote access request."""
        # Validate against policy
        violations = []

        if request.duration_hours > self.policy["max_session_hours"]:
            violations.append(
                f"Duration {request.duration_hours}h exceeds maximum {self.policy['max_session_hours']}h"
            )

        for target in request.target_systems:
            for prohibited in self.policy["prohibited_targets"]:
                pattern = prohibited.replace("*", "")
                if target.startswith(pattern):
                    violations.append(f"Target {target} is in prohibited list (safety systems)")

        if violations:
            print(f"[!] Policy violations found:")
            for v in violations:
                print(f"    - {v}")
            return ""

        self.access_requests[request.id] = request
        print(f"[+] Access request {request.id} submitted for approval")
        return request.id

    def list_pending_requests(self):
        """List all pending access requests for approval."""
        pending = [r for r in self.access_requests.values()
                   if r.status == AccessRequestStatus.PENDING]

        print(f"\n{'='*65}")
        print("PENDING OT REMOTE ACCESS REQUESTS")
        print(f"{'='*65}")

        if not pending:
            print("  No pending requests")
            return

        for req in pending:
            print(f"\n  Request: {req.id}")
            print(f"    Requestor: {req.requestor} ({req.requestor_type})")
            print(f"    Purpose: {req.purpose}")
            print(f"    Targets: {', '.join(req.target_systems)}")
            print(f"    Duration: {req.duration_hours} hours")
            print(f"    Start: {req.requested_start}")
            print(f"    Submitted: {req.created}")

    def generate_audit_report(self):
        """Generate audit report of all remote access activity."""
        print(f"\n{'='*65}")
        print("OT REMOTE ACCESS AUDIT REPORT")
        print(f"{'='*65}")
        print(f"Report Date: {datetime.now().isoformat()}")
        print(f"Total Requests: {len(self.access_requests)}")

        status_counts = {}
        for req in self.access_requests.values():
            status = req.status.value
            status_counts[status] = status_counts.get(status, 0) + 1

        print(f"\nRequest Status:")
        for status, count in status_counts.items():
            print(f"  {status}: {count}")

        print(f"\nDetailed Audit Trail:")
        for req in self.access_requests.values():
            print(f"\n  {req.id} - {req.requestor} ({req.requestor_type})")
            print(f"    Status: {req.status.value}")
            print(f"    Targets: {', '.join(req.target_systems)}")
            for entry in req.audit_trail:
                print(f"    [{entry['timestamp']}] {entry['action']} by {entry['actor']}")


if __name__ == "__main__":
    manager = OTConduitManager()

    # Vendor access request
    vendor_req = RemoteAccessRequest(
        requestor="John Smith - Siemens Field Service",
        requestor_type="vendor",
        purpose="Annual PLC firmware update for S7-1500 controllers",
        target_systems=["PLC-REACTOR-01", "PLC-REACTOR-02"],
        duration_hours=3,
        requested_start="2025-03-15T08:00:00",
    )

    req_id = manager.submit_request(vendor_req)
    if req_id:
        manager.list_pending_requests()

    # Simulate approval
    manager.access_requests[req_id].approve(
        approver="Plant Manager - Jane Doe",
        conditions="OT engineer must shadow the vendor session",
    )

    manager.generate_audit_report()

Key Concepts

Term Definition
Conduit IEC 62443 controlled communication path between security zones with defined security policies
Jump Server Hardened intermediary server in the DMZ through which all remote OT access must transit
Session Recording Capture of all screen activity, keystrokes, and commands during a remote access session for audit
Approval-Based Access Workflow requiring plant operations manager authorization before remote access credentials are activated
Vendor Escort Practice of having an internal OT engineer monitor vendor remote sessions in real time
Break-Glass Access Emergency access procedure bypassing normal approval workflow for critical situations

Output Format

OT REMOTE ACCESS CONDUIT REPORT
==================================
Date: YYYY-MM-DD

CONDUIT STATUS:
  Internal Access Conduit: [Active/Inactive]
  Vendor Access Conduit: [Active/Inactive]

ACCESS REQUESTS (Last 30 Days):
  Submitted: [count]
  Approved: [count]
  Rejected: [count]
  Average Session Duration: [hours]

POLICY COMPLIANCE:
  MFA Enforcement: [100%]
  Session Recording: [100%]
  Time-Limited Sessions: [compliance %]
  Prohibited Target Attempts: [count blocked]

04 同仓库更多 Skills

mukul975
mukul975/Anthropic-Cybersecurity-Skills

verifying-build-provenance-with-slsa-sigstore

Verifies artifact signatures and SLSA provenance using Sigstore's cosign (verify, verify-attestation, verify-blob-attestation) and slsa-verifier (verify-artifact), enforcing keyless OIDC builder identity and source repo against SLSA Build levels. Use in CI/CD before deploying artifacts, when consuming third-party attestations, establishing a SLSA Build L3 pipeline, or confirming provenance during incident response or admission control.

Skill与智能体开发
★ 3.6w
mukul975
mukul975/Anthropic-Cybersecurity-Skills

validating-tpm-measured-boot-attestation

Verifies TPM 2.0 measured-boot integrity and remote attestation with tpm2-tools -- reading PCRs (tpm2_pcrread), replaying the boot event log, generating and checking signed quotes (tpm2_quote/tpm2_checkquote), and sealing secrets to a PCR policy. Use to confirm a system booted trusted firmware/kernel for Zero Trust device posture, detect boot-chain tampering via PCR drift, or build a golden-value baseline for fleet attestation.

Skill与智能体开发
★ 3.6w
mukul975
mukul975/Anthropic-Cybersecurity-Skills

validating-backup-integrity-for-recovery

Validates backup integrity through cryptographic hash verification, automated restore testing, corruption detection, and recoverability checks to confirm backups are reliable for disaster recovery and ransomware response. Use before relying on backups for recovery, when building post-backup validation pipelines, auditing backup infrastructure for compliance, or checking immutable/air-gapped backups for silent corruption or tampering.

测试与质量Skill与智能体开发
★ 3.6w
mukul975
mukul975/Anthropic-Cybersecurity-Skills

triaging-windows-with-kape

Runs KAPE (Kroll Artifact Parser and Extractor) to collect targeted forensic artifacts (registry hives, $MFT, event logs, prefetch, browser data) via Targets and parse them with Modules wrapping Eric Zimmerman's EZ Tools (PECmd, MFTECmd, RECmd). Use during early incident containment/triage when full disk imaging is impractical but a defensible, parseable Windows artifact set is needed quickly, including at-scale remote collection.

Skill与智能体开发
★ 3.6w
mukul975
mukul975/Anthropic-Cybersecurity-Skills

triaging-vulnerabilities-with-ssvc-framework

Triages and prioritizes vulnerabilities with CISA's Stakeholder-Specific Vulnerability Categorization (SSVC) decision tree, weighing exploitation status (via the CISA KEV catalog and FIRST EPSS API), technical impact, automatability, and mission prevalence to output Track/Track*/Attend/Act decisions. Use when prioritizing vulnerability scan results (OpenVAS, Nessus, Qualys) for remediation planning beyond raw CVSS scores.

Skill与智能体开发
★ 3.6w
mukul975
mukul975/Anthropic-Cybersecurity-Skills

triaging-security-incident

Performs initial triage of security incidents using the NIST SP 800-61r3 and SANS PICERL frameworks, classifying incident type, assigning priority by business impact, and routing to the appropriate response team. Use when a SIEM/EDR alert needs human classification, concurrent alerts must be prioritized, or a user report or threat-intel IOC match requires initial incident categorization.

Skill与智能体开发
★ 3.6w