34 min read

Secure Remote Access for MSPs: Configuration Checklist

Secure Remote Access for MSPs: Configuration Checklist

Secure Remote Access for MSPs: A Practical Configuration Checklist

Remote access allows managed service providers (MSPs) to troubleshoot endpoints, resolve technical issues, maintain systems, and support clients without sending technicians on-site. However, the same capabilities that make remote management efficient can create security risks when access is poorly configured.

A compromised technician account, an exposed remote desktop service, or excessive administrative permissions can give attackers access to sensitive client environments. For MSPs managing multiple organizations, the consequences can extend beyond a single device.

Secure remote access requires more than encrypted connections and strong passwords. MSPs must control technician identities, restrict administrative privileges, protect managed endpoints, monitor remote sessions, and establish procedures for responding to suspicious activity.

This guide provides a practical remote access security checklist for MSPs, including specific configuration steps, verification methods, and operational controls that help protect both technicians and client systems.

What Is Secure Remote Access for MSPs?

Secure remote access is the practice of allowing authorized technicians to connect to and manage remote systems while enforcing authentication, access restrictions, encryption, endpoint protection, and activity monitoring.

For MSPs, remote access commonly involves several technologies:

  • Remote monitoring and management (RMM) platforms: Centralized tools for endpoint administration, troubleshooting, scripting, automation, and remote support.
  • Remote desktop software: Applications that provide interactive control of a remote computer.
  • Virtual private networks (VPNs): Encrypted network connections that provide access to approved internal resources.
  • Zero Trust Network Access (ZTNA): Access solutions that evaluate identity and other defined conditions before granting access to specific applications or resources.
  • Privileged access management (PAM): Controls that govern elevated credentials, administrative sessions, and access duration.

These technologies serve different purposes and may be used together. Regardless of the connection method, secure access depends on controlling who can connect, which resources they can reach, what actions they can perform, and how their activities are recorded.

The National Institute of Standards and Technology (NIST) recommends protecting the entire remote access environment, including access infrastructure, connecting devices, authentication systems, and the information transmitted between them.

For MSPs, these protections must also account for separation between customers.

Why Remote Access Security Requires Special Attention from MSPs

MSPs often manage hundreds or thousands of endpoints across multiple organizations. Their technicians may have administrative privileges, access to sensitive business information, and tools capable of executing commands across entire device groups.

This creates several important security concerns.

Compromised Technician Accounts

An attacker who obtains a technician's credentials may be able to access managed devices, execute commands, or modify endpoint configurations, depending on the account's permissions.

Multifactor authentication and least-privilege access help reduce this risk, but neither eliminates it entirely.

Unauthorized Remote Management Tools

Attackers can abuse legitimate remote access applications to establish persistence, interact with compromised systems, or bypass security controls that primarily detect conventional malware.

The joint Guide to Securing Remote Access Software, published by CISA, NSA, FBI, MS-ISAC, and Israel's National Cyber Directorate, describes how attackers misuse legitimate remote management tools and recommends restricting unauthorized applications, monitoring suspicious activity, and strengthening access controls.

Excessive Cross-Client Privileges

Technicians sometimes receive broad access to simplify support operations.

However, an account that can administer every customer environment creates a substantially larger potential impact if compromised.

Privileged Script Execution and Automation

RMM platforms can execute administrative scripts and automated actions across numerous endpoints.

When these capabilities are poorly restricted, an attacker with sufficient privileges may be able to distribute malicious commands, modify security configurations, or disrupt multiple managed systems.

Insufficient Activity Monitoring

Without reliable logs, MSPs may struggle to determine which technician accessed a device, when the connection occurred, or whether sensitive administrative actions were performed.

Unsecured Endpoints and Remote Access Infrastructure

Unpatched systems, exposed services, weak workstation security, and poorly maintained remote access gateways can undermine otherwise strong authentication controls.

These concerns make secure remote access an operational responsibility, not simply a software feature.

Secure Remote Access Configuration Checklist for MSPs

The following checklist covers the core controls MSPs should configure and verify. Each item includes the security objective, implementation considerations, and a practical method for confirming that the control works.

The Critical and High priorities represent suggested implementation priorities based on the potential security impact. They are not official NIST or CISA severity classifications. Organizations should adjust priorities according to their own risk assessments and customer requirements.

1. Enforce Multifactor Authentication for Every Technician

Security objective: Prevent passwords alone from granting access to remote management systems.

  • Require MFA for all technician and administrator accounts.
  • Prefer phishing-resistant authentication methods where supported.
  • Apply MFA to VPNs, identity providers, remote support tools, and other administrative entry points.
  • Verify MFA enforcement for technicians using SSO.
  • Disable or tightly control authentication exceptions.
  • Establish secure procedures for MFA enrollment, replacement, and account recovery.
  • Maintain a protected emergency administrative access procedure.
  • Monitor unexpected authentication prompts, failed attempts, and unusual enrollment changes.

Configuration guidance

Enable mandatory MFA at the organization level rather than relying on individual technicians to opt in.

Where supported, use phishing-resistant authentication such as FIDO2 security keys or appropriately configured passkeys. Authenticator applications provide another layer of protection when phishing-resistant options are unavailable.

SMS-based authentication should generally be treated as a fallback rather than the preferred option.

Account recovery deserves particular attention. An attacker should not be able to bypass MFA simply by requesting a password reset or registering a new authentication device.

For example, Level administrators can enforce authenticator-based two-factor authentication through Settings → Security → Require two-factor authentication.

Once enabled, technicians using native Level authentication without configured 2FA must complete enrollment before they can sign in.

Single sign-on requires separate MFA verification.

When a Level technician account uses SSO, its native Level 2FA settings are bypassed. Authentication and MFA enforcement become the responsibility of the connected identity provider.

An MSP using Microsoft Entra, Google Workspace, or another supported OIDC provider must therefore confirm that MFA is required by the identity provider's applicable authentication policies.

Enabling Level's native 2FA requirement alone should not be treated as proof that SSO accounts are protected by a second factor.

Plan for identity-provider outages.

Level's documentation recommends maintaining a protected emergency administrator account using native authentication outside the SSO dependency.

This account should have strong credentials, mandatory 2FA, tightly controlled access, and a documented emergency-use procedure. MSPs should periodically verify that it remains accessible without using it for routine administration.

How to verify

Test an account without enrolled MFA and confirm that it cannot complete an ordinary native sign-in until enrollment requirements are satisfied.

For SSO accounts, verify the identity provider's policy and test the sign-in flow to confirm that the required authentication method is enforced.

Also verify that recovery and enrollment procedures require appropriate identity validation.

Test the emergency administrator account through an approved recovery exercise without weakening normal authentication requirements.

Pass condition: Every tested technician account satisfies the required authentication policy, and a protected emergency access method remains available when the identity provider cannot be used.

Priority: Critical

2. Apply Role-Based Access Control and Least Privilege

Security objective: Give technicians only the access necessary for their assigned responsibilities.

  • Create separate roles for technicians, senior administrators, and organization owners.
  • Restrict technicians to their assigned customers and device groups.
  • Separate routine support permissions from high-risk administrative capabilities.
  • Limit access to remote control, background management, scripting, and automation.
  • Restrict command-line interface access to authorized roles.
  • Inventory API credentials and restrict their privileges.
  • Avoid shared administrative accounts.
  • Review elevated privileges regularly.
  • Remove permissions promptly when responsibilities change.

Configuration guidance

A technician performing routine workstation troubleshooting may need interactive remote control and access to basic device information. That does not automatically justify permission to modify organization-wide security settings or execute privileged scripts across every customer.

MSPs should distinguish between three levels of authority:

  • Routine support: Access to assigned devices and approved troubleshooting functions.
  • Elevated technical support: Additional administrative capabilities needed for complex remediation.
  • Platform administration: Authority to manage security settings, roles, and organization-wide resources.

Client separation is especially important. Permissions should be scoped to the appropriate customer environment rather than granted globally for convenience.

In Level, administrators can create custom roles under Workspace → Permissions.

The permission model distinguishes between:

  • Management permissions for resources such as scripts, automations, and monitor policies.
  • Device permissions for remote control and background management.
  • Group access for devices, scripts, and automations.
  • CLI access permissions that determine whether technicians can use Level's command-line interface.

Script management permission is different from script execution access. A technician may be permitted to run scripts from an authorized script group without having permission to create or modify those scripts.

MSPs should also review group inheritance. In Level, the Include future groups option can automatically extend a role's access to newly created subgroups. This is useful for administration but may unintentionally expand permissions as new customer environments are added.

Administrator status requires particular attention.

Level administrators bypass all role-based permission checks. Administrator status is a workspace-level account property managed through Workspace → Team, not an ordinary custom role.

An MSP should therefore avoid assigning administrator status to technicians who are expected to remain restricted to specific customers or device groups.

CLI and API access require separate controls.

Level's CLI provides terminal-based access to supported management functions. Administrators always have CLI access, while other technicians require the appropriate role permission.

API keys are managed separately under Settings → API keys. They can be configured for read-only or read-and-write access.

MSPs should create identifiable keys for individual integrations, use read-only access unless write operations are required, and store credentials securely.

API keys should not be treated as ordinary technician role assignments. Their access level and lifecycle need separate review.

For highly sensitive administrative operations, consider time-limited privilege elevation through a compatible privileged access management solution.

How to verify

Use a test account assigned to a restricted role.

Confirm that the account can access its authorized devices but cannot view or operate on resources belonging to unrelated customers.

Test remote control, background management, script execution, CLI sign-in, and administrative configuration separately.

Also confirm that restricted technician accounts have not inadvertently been granted administrator status.

Review active API keys and verify that each has an identifiable owner, documented purpose, and appropriate access level.

Pass condition: Restricted accounts can perform approved responsibilities but cannot access unauthorized resources or administrative functions, while CLI and API credentials remain independently controlled.

Priority: Critical

3. Restrict Where Remote Connections Can Originate

Security objective: Reduce exposure to unauthorized networks and access locations.

  • Identify approved remote management entry points.
  • Restrict technician sign-ins to trusted networks where operationally appropriate.
  • Use IP allowlists for environments with predictable network locations.
  • Require approved VPN or ZTNA access when the management architecture calls for it.
  • Close unnecessary inbound ports.
  • Remove direct internet exposure from administrative services that do not require it.
  • Review exceptions for vendors, contractors, and emergency access.

Configuration guidance

An IP allowlist can restrict administrative access to approved office networks or known VPN egress addresses.

However, IP restrictions do not replace identity verification. Shared networks, compromised trusted devices, and changing network addresses introduce limitations.

Administrators must also take precautions when enabling restrictions.

For example, Level's Settings → Security → IP Allowlist allows organizations to define trusted IP addresses.

Add the administrator's current IP address before enabling the allowlist. Otherwise, the administrator may immediately lose access to the platform.

For technicians working remotely, MSPs should consider whether IP restrictions are operationally suitable or whether access should be routed through an approved network with a predictable egress address.

Traditional remote desktop services deserve special attention.

Avoid exposing Windows Remote Desktop Protocol (RDP) directly to the public internet. Where RDP is necessary, place access behind an appropriately secured gateway, VPN, or other access-control architecture, and apply MFA, monitoring, and network restrictions.

Agent-based remote support platforms may use outbound connections rather than requiring technicians to expose an inbound RDP port. This can simplify network configuration, although the management platform and technician accounts still require strong security.

How to verify

Attempt to access the management interface from an unauthorized network and confirm that the configured restriction blocks the connection.

Separately verify that legitimate technicians can still connect through approved access paths.

Check externally exposed services using authorized network scanning and firewall configuration reviews.

Pass condition: Unauthorized network paths are blocked without preventing approved administrative access.

Priority: Critical

4. Secure Technician Workstations

Security objective: Prevent compromised technician devices from becoming entry points into customer systems.

  • Use managed, organization-approved workstations.
  • Enable full-disk encryption.
  • Keep operating systems, browsers, and remote support applications updated.
  • Deploy appropriate endpoint detection and protection controls.
  • Enable host firewalls.
  • Remove unnecessary local administrator privileges.
  • Require automatic workstation locking.
  • Separate administrative activities from routine personal browsing.
  • Define rules for using personal or unmanaged devices.

Configuration guidance

A technician workstation is part of the MSP's privileged access infrastructure.

Even strong MFA may not fully protect a customer environment when an attacker controls an already-authenticated workstation or steals active session information.

MSPs should therefore apply stricter standards to devices used for administrative access than to ordinary employee endpoints.

For higher-risk environments, consider dedicated privileged-access workstations and device-compliance policies that restrict management access from untrusted devices.

These policies may require capabilities provided by an identity provider, endpoint management system, or other security platform. They should not be assumed to exist in every RMM application.

How to verify

Review workstation compliance reports and confirm that required security controls are active.

Where device-based restrictions have been implemented, test whether an unmanaged or noncompliant workstation can access administrative tools.

Pass condition: Approved technician workstations meet the defined security baseline, and noncompliant devices are restricted wherever device-compliance enforcement is required.

Priority: Critical

5. Protect Remote Sessions and Sensitive Data

Security objective: Prevent unauthorized interception, unnecessary disclosure, and misuse of data during remote support.

  • Require encrypted remote connections.
  • Use supported, maintained remote access software.
  • Disable insecure protocols and obsolete encryption configurations.
  • Restrict clipboard access and file transfer when not operationally necessary.
  • Establish rules for handling customer credentials and confidential files.
  • Configure session inactivity controls.
  • End remote sessions when troubleshooting is complete.
  • Avoid storing credentials in scripts, tickets, or unsecured documents.

Configuration guidance

Encryption protects data in transit, but it does not determine whether a technician should have access to that data.

For example, a fully encrypted remote session can still expose confidential documents if the technician has unrestricted access to a workstation.

MSPs should combine encrypted sessions with appropriate permissions, customer-specific policies, and controls over sensitive actions.

Clipboard synchronization and file transfer are useful support capabilities. They also create potential data movement paths that should be governed according to customer requirements.

Credentials used during support should be stored in approved credential-management systems rather than copied into ordinary ticket notes.

Session inactivity controls provide another layer of protection.

For example, Level administrators can configure the application inactivity timeout under Settings → Security, with available durations ranging from 15 minutes to 7 days. The documented default is 24 hours.

This timeout applies to browser inactivity within Level. It should not be confused with an operating system screen-lock policy or assumed to terminate every existing endpoint connection instantly.

Organizations should select a timeout appropriate to their security policies and operating requirements.

How to verify

Review the remote access product's documented encryption architecture and supported security settings.

Test inactivity behavior, sensitive capability restrictions, and session termination.

Verify what happens to an existing connection when the technician signs out or loses application access.

Pass condition: Remote connections use the expected encryption, configured restrictions are enforced, and session expiration behaves according to documented policy.

Priority: High

Security objective: Ensure remote sessions follow the customer's authorization and privacy requirements.

  • Define which sessions require end-user approval.
  • Configure visible connection notifications where appropriate.
  • Require explicit consent for sensitive or privacy-sensitive workstations.
  • Establish separate procedures for unattended servers.
  • Document authorized after-hours maintenance.
  • Control how unanswered approval prompts are handled.
  • Distinguish interactive remote control from background management.
  • Ensure customer contracts define authorized access expectations.

Configuration guidance

Not every remote connection requires the same approval workflow.

Attended employee workstations may benefit from explicit user authorization, while unattended servers may require preapproved administrative access without an interactive user.

For sensitive environments, the distinction between notifying a user and requiring consent is important.

A notification informs someone that access is occurring. An approval requirement prevents the session from starting until the necessary authorization is received.

MSPs must also determine what happens when an approval prompt receives no response.

For example, Level provides two remote control approval modes:

  • Notify end user: A notification appears when the remote control session begins.
  • Ask for approval: The end user receives an approval prompt before the screen-sharing session begins.

When approval is requested, Level also provides two timeout behaviors:

  • Connect if the end user doesn't answer: The connection proceeds after the timeout.
  • Require approval: The connection is denied unless the end user explicitly approves.

For sensitive workstations requiring affirmative consent, the second timeout option is the appropriate configuration.

Level's organization-wide default is configured under Settings → Security, while individual device groups can override the default through Group Settings → Security.

Administrators should review these overrides before applying changes across the organization. Forcing an organization-wide approval configuration onto descendant groups can overwrite existing group-specific settings.

Important: Remote control approval is not the same as background management authorization.

In Level, end-user approval applies to interactive screen-sharing sessions. Background management tools, including terminal, file explorer, processes, and services, operate without the same approval prompt or screen-sharing notification.

This distinction matters because a technician may be authorized to perform background operations even when an interactive remote control session would require user approval.

MSPs must therefore govern background access separately through appropriate roles, device-group permissions, activity monitoring, and customer authorization procedures.

How to verify

Initiate a test remote control session against a device configured to require approval.

Confirm that access cannot begin without affirmative authorization, including when the user ignores the prompt.

Next, separately test background management using a restricted technician role.

Verify that the role cannot access background tools unless that access has been explicitly granted.

Pass condition: Interactive remote control follows the configured consent policy, and background management permissions independently enforce the customer's access requirements.

Priority: High

7. Enable Remote Access Logging and Audit Trails

Security objective: Establish reliable evidence of remote activity and support incident investigation.

  • Record technician sign-ins and authentication failures where the authentication system supports them.
  • Capture completed remote session activity.
  • Associate sessions with identifiable technician accounts.
  • Record device and customer context.
  • Log significant administrative actions where supported.
  • Monitor scripting, command execution, and configuration changes using appropriate telemetry.
  • Distinguish successful sessions from failed or rejected access attempts.
  • Forward relevant logs to centralized monitoring systems where supported.
  • Protect logs from unauthorized modification or deletion.
  • Define retention requirements based on contracts, regulations, and investigative needs.

Configuration guidance

Remote access logs should make it possible to answer several fundamental questions:

Who accessed the device? Which device was involved? When did the connection occur? What relevant actions were performed? Was the activity authorized?

The answer may require combining multiple data sources, including RMM activity logs, identity-provider logs, endpoint telemetry, and service desk records.

MSPs should not assume every remote access product records screen contents or every action performed inside a remote desktop session.

Session metadata, administrative activity logs, command-level records, and screen recordings are different forms of evidence.

For example, Level's Activity documentation explains that completed remote control and terminal sessions generate activity records containing technician identity, device information, session type, connection type, and duration.

However, remote access Activity entries do not include terminal command text or output.

These records can establish that a terminal session occurred, but they do not independently establish every command executed during that session.

Where command-level investigation is required, MSPs should use additional endpoint auditing or other appropriate security telemetry.

Level also records configuration changes, script modifications, automation activity, and other administrative events. Its Activity records are read-only and currently retained indefinitely, according to its published documentation.

Completed sessions and failed access attempts require different validation.

A completed Level remote control or terminal session produces a remote access Activity event. A failed sign-in or denied connection should not automatically be expected to produce the same type of record.

MSPs should identify which logs capture authentication failures, rejected access attempts, and completed sessions, then verify each event type in the appropriate system.

The joint Guide to Securing Remote Access Software recommends that MSPs and their customers maintain effective monitoring and logging while protecting log systems from unauthorized access or alteration.

How to verify

Perform approved test remote control and terminal sessions, then end them and confirm that the expected completed-session records appear.

Separately test authentication failures and rejected connections, checking the appropriate identity-provider, application, or endpoint logs.

Perform an approved administrative configuration change and verify its audit record.

For environments requiring command-level evidence, confirm that the complementary monitoring system captures the necessary information.

Pass condition: Completed sessions and administrative changes generate their documented records, authentication failures are visible in the appropriate monitoring systems, and the organization understands which actions are not captured by the RMM activity log.

Priority: Critical

8. Harden Managed Endpoints and Remote Access Infrastructure

Security objective: Prevent vulnerable managed systems and access components from undermining remote support security.

  • Maintain accurate inventories of managed endpoints.
  • Apply operating system and application updates.
  • Patch VPN gateways and remote access appliances promptly.
  • Enable endpoint firewalls and appropriate malware protection.
  • Disable unused remote access services.
  • Remove obsolete remote management agents.
  • Identify unsupported operating systems.
  • Review exposed administrative ports and services.
  • Use network segmentation to restrict unnecessary access between systems.
  • Maintain appropriate backups and recovery procedures.

Configuration guidance

Remote access security depends on the condition of both the technician's device and the managed endpoint.

An up-to-date RMM platform cannot compensate for every weakness in an unmanaged operating system, unpatched VPN appliance, or exposed administrative service.

MSPs should identify remote access infrastructure as a priority for vulnerability management.

Systems that provide administrative entry into multiple environments deserve especially careful patching and monitoring.

Endpoint inventories should also identify which remote access applications are installed. Unauthorized or forgotten remote management tools create additional paths that may bypass established security procedures.

For broader endpoint maintenance, combine this checklist with a structured patch management program.

How to verify

Compare inventory records against actual endpoint and software discovery results.

Review vulnerability scans, patch status, and exposed service reports.

Confirm that obsolete agents and unnecessary administrative services have been removed.

Pass condition: Managed endpoints meet the applicable security baseline, critical remote access infrastructure is maintained, and unnecessary remote access services have been removed or formally approved as exceptions.

Priority: Critical

9. Detect Suspicious Remote Access and Administrative Activity

Security objective: Identify unauthorized access or misuse before it spreads across customer environments.

  • Alert on unusual administrative sign-ins.
  • Investigate repeated authentication failures.
  • Monitor unexpected MFA changes.
  • Review technician activity outside established working patterns.
  • Detect unexpected scripts or commands executed across multiple endpoints.
  • Monitor new administrative accounts and privilege changes.
  • Identify unauthorized remote access software.
  • Correlate RMM activity with endpoint and identity security events.
  • Establish authorization requirements for high-impact bulk actions.

Configuration guidance

Not all suspicious activity is obviously malicious.

An administrator connecting outside business hours may be performing scheduled maintenance. A technician accessing several devices may be resolving a widespread outage.

The objective is to identify behavior that falls outside expected patterns and verify whether it has a legitimate operational explanation.

For example, a technician account that suddenly launches administrative scripts across multiple unrelated customers should receive greater scrutiny than a routine connection to an assigned workstation.

MSPs should establish controls around particularly sensitive actions, including:

  • Executing scripts across large endpoint groups.
  • Modifying security monitoring or endpoint protection settings.
  • Changing automation schedules or execution targets.
  • Distributing software to multiple customer environments.
  • Modifying administrative roles or device-group permissions.

Where supported, these operations should be restricted to appropriately authorized roles.

For higher-impact changes, MSPs can establish internal approval procedures even when the RMM platform itself does not provide a native approval workflow for every action.

Script and automation permissions should also be reviewed separately.

A technician authorized to run a previously approved maintenance script does not necessarily need permission to create new scripts or modify automation workflows.

The script's execution target matters as much as the technician's permission to run it.

For example, a script designed to restart a workstation service may be appropriate for a designated group of office computers but unsuitable for production servers running critical applications.

MSPs should therefore verify that a technician has permission to execute a script and that the intended target devices are appropriate for the operation.

Monitoring can be improved by correlating remote activity with approved tickets, maintenance windows, and authorization records.

How to verify

Generate approved test events for selected detection rules.

Confirm that relevant monitoring systems receive the events, produce the expected alerts, and route them to the correct personnel.

Use a restricted test account to confirm that prohibited bulk operations cannot be performed.

Also verify that script execution is restricted to authorized device groups.

Pass condition: Unauthorized high-impact operations are prevented by access controls, script execution remains within approved scope, and the monitoring environment detects selected suspicious activity according to documented rules.

Priority: High

10. Establish Access Revocation and Incident Response Procedures

Security objective: Quickly contain compromised access and remove permissions that are no longer required.

  • Define a process for disabling compromised technician accounts.
  • Revoke available authentication sessions and tokens.
  • Remove access when employees leave or change responsibilities.
  • Revoke unused or compromised API keys.
  • Review and terminate unauthorized CLI sessions.
  • Rotate exposed credentials and secrets.
  • Establish escalation procedures for suspected unauthorized remote access.
  • Document customer notification responsibilities.
  • Maintain emergency administrative access procedures.
  • Test account revocation and recovery workflows.

Configuration guidance

Disabling an account does not always invalidate every existing session immediately.

MSPs must distinguish between several forms of access:

  • Identity-provider sessions: Authentication sessions maintained by the organization's SSO or identity provider.
  • Application sessions: Authenticated browser or command-line sessions associated with the RMM platform.
  • Active remote management connections: Connections used to interact with managed endpoints.
  • API credentials: Keys used by external integrations or scripts to access management functions.
  • Endpoint-level credentials: Local or domain credentials that may remain usable independently of the RMM account.

Each may require a different revocation or containment action.

For example, Level's Settings → Sessions page allows a technician to review and sign out their own authenticated browser and CLI sessions.

However, this page is personal. It does not provide an organization-wide view of every other technician's sessions.

Level's documentation also specifies that signing out a session prevents future access renewal, which occurs within an hour at most. This should not be represented as guaranteed instantaneous termination of every existing connection.

Disabling CLI access for a role blocks new CLI sign-ins and access renewal, but an already-issued access token can remain valid until it expires, for up to one hour.

API keys require independent revocation.

Level administrators can review keys under Settings → API keys, identify their access levels, and delete keys that should no longer be authorized.

Deleting a Level API key revokes it immediately. Integrations attempting to use that key will receive authentication errors on subsequent requests.

Where a legitimate integration must remain operational, MSPs should establish a replacement key and update the integration before revoking its existing credential, unless immediate containment takes priority.

Signing out of Level also does not shut down its installed agents. Managed devices remain online and monitored.

MSPs should therefore avoid relying on a single sign-out operation as their entire account-compromise response.

A practical response includes disabling the affected identity, revoking available sessions and API credentials, resetting exposed secrets, reviewing recent administrative activity, and evaluating whether managed endpoints require additional containment.

The scope of containment should be based on evidence and the potential impact across customer environments.

How to verify

Conduct a controlled test using an authorized technician account.

Disable its access, revoke the available sessions, and verify that the account cannot establish new connections.

Observe whether previously authenticated sessions remain usable during the documented token or session lifetime.

For API credentials, use a dedicated test key to verify that revocation blocks subsequent authenticated API requests.

Confirm that security personnel can retrieve relevant activity records and follow the incident response procedure.

Pass condition: New access is blocked, existing sessions are handled according to documented behavior, compromised API credentials are independently revoked, and the MSP has a tested process for investigating residual access.

Priority: Critical

How to Validate a Secure Remote Access Configuration

A checklist is useful only when the organization can demonstrate that the controls are functioning.

MSPs should distinguish between three levels of validation:

  1. Configured: The security setting is enabled or documented.
  2. Enforced: Testing confirms that the setting prevents an unauthorized action.
  3. Monitored: The organization can detect or investigate relevant activity when the control is used or challenged.

For example, enabling MFA is a configuration step. Testing whether an unenrolled account is blocked verifies enforcement. Monitoring suspicious authentication activity provides the corresponding visibility.

The same principle applies to permissions, network restrictions, remote control approval, and administrative logging.

A Practical Validation Scenario

Consider an MSP supporting three customers with different security requirements.

Customer A has ordinary office workstations. Customer B operates sensitive financial systems. Customer C maintains unattended infrastructure.

The MSP assigns one technician to Customer A and another to Customer B.

During validation, the MSP tests whether:

  • The Customer A technician can access only approved Customer A devices.
  • The Customer A technician cannot connect to Customer B or C.
  • Customer B workstations require explicit approval for interactive remote control.
  • Customer B background management access is separately restricted according to its security policy.
  • Customer C servers permit only the authorized unattended access workflow.
  • Both technicians must satisfy mandatory authentication requirements, including any identity-provider MFA requirements for SSO users.
  • Neither technician has unnecessary administrator status.
  • The technicians cannot execute scripts against unauthorized customer groups.
  • Restricted technicians cannot use the CLI unless their role explicitly permits it.
  • Active API keys have documented purposes and appropriate access levels.
  • Completed remote control and terminal sessions generate their documented activity records.
  • Failed authentication and rejected access attempts can be investigated using the appropriate logging systems.
  • Disabling a technician's access prevents new sessions and triggers the documented procedures for handling existing sessions.

This exercise exposes practical problems that may not be visible in configuration screens.

For instance, a technician might have a restricted role but retain access through inherited device-group permissions.

Alternatively, a group-level remote approval override might allow unattended connections that the organization's general policy intended to block.

An administrator might also discover that access to a script group grants execution of scripts that were not intended for that technician's responsibilities.

Validation should include both permitted and denied actions.

Each failed authorization test should produce a clear remediation task rather than being treated as an acceptable configuration difference.

Common Remote Access Security Mistakes MSPs Should Avoid

Even mature IT teams can create unnecessary exposure through operational shortcuts.

Using Shared Administrator Accounts

Shared accounts make it harder to establish accountability and determine which individual performed an action.

Give technicians identifiable accounts and use approved privileged credential management where shared system credentials remain operationally necessary.

Assuming a VPN Makes All Remote Access Secure

A VPN protects the connection according to its configuration, but it does not automatically verify that every action or destination is authorized.

Access still requires identity controls, endpoint protection, appropriate network restrictions, and monitoring.

Granting Broad Permissions to Save Time

Excessive access makes routine administration convenient but increases the potential impact of compromised accounts.

Use customer-specific roles and elevated access only when the work requires it.

Treating Remote Control Approval as Universal Access Approval

An end-user consent prompt may govern screen sharing without restricting background terminal access, scripting, or file management.

MSPs must verify which operations require approval and which are controlled through role-based permissions.

Allowing Unapproved Remote Support Applications

An unauthorized remote tool may provide an alternate management path outside the MSP's normal controls and audit process.

Maintain an approved software inventory and investigate unexpected remote access applications.

Allowing Unrestricted Script Execution

Scripts can make routine maintenance efficient, but broad execution privileges increase the potential damage from a compromised technician account.

Separate script creation, management, and execution privileges wherever the platform permits.

Treating Audit Logs as a Substitute for Active Monitoring

Logs are valuable for investigations, but an unattended log repository may not provide timely detection.

Define which events require alerting and assign responsibility for reviewing them.

Ignoring Technician Offboarding

Former employees and contractors should not retain access through forgotten local accounts, backup credentials, API keys, or individual remote support platforms.

Integrate remote management access into a documented identity offboarding process.

Building an Ongoing Remote Access Security Review Process

Remote access configuration is not a one-time task.

MSPs continuously add customers, onboard technicians, deploy endpoints, change permissions, and introduce new software. Each operational change can affect the security controls that were previously validated.

A sustainable review process should include the following activities.

During Technician Onboarding

Confirm that the technician receives an individual account, approved authentication methods, appropriate device-group permissions, and access through a compliant workstation.

For SSO users, verify the identity provider's MFA policies before granting administrative access.

When Adding a New Customer

Document authorized connection methods, customer-specific approval requirements, administrative boundaries, logging expectations, and incident notification procedures.

Test that the new environment remains separated from existing customers.

After Significant Configuration Changes

Revalidate affected permissions, session controls, authentication rules, and network restrictions.

Changes to inherited roles, future-group access, script permissions, CLI permissions, or remote control approval overrides deserve particular attention because their effects may extend beyond one technician.

During Periodic Security Reviews

Review privileged accounts, unused remote access tools, unusual connection activity, authentication exceptions, API credentials, and endpoint vulnerabilities.

The review frequency should reflect customer risk, contractual obligations, and how often the environment changes.

After a Security Incident

Determine whether existing access restrictions worked as intended, whether monitoring provided sufficient evidence, and whether the incident exposed gaps in account management or response procedures.

Use the findings to update configuration standards and test procedures.

How Level.io Supports Secure Remote Access for MSPs

Remote access security becomes easier to manage when technicians can perform their work through a centralized platform with consistent access controls and activity visibility.

Level.io combines remote monitoring and management with browser-based remote control, background management, role-based permissions, security settings, and administrative activity logging.

Several capabilities are particularly relevant to an MSP's remote access configuration checklist.

Centralized Authentication and Access Restrictions

Level supports mandatory two-factor authentication for native accounts, role-based permissions, device-group access restrictions, and IP allowlisting.

MSPs can use these controls to limit which technicians can sign in and which customer endpoints they can manage.

Level also supports single sign-on with preconfigured integrations for Microsoft Entra and Google Workspace, along with a generic OpenID Connect (OIDC) integration for compatible identity providers such as Okta.

Only one identity provider can be connected at a time. Connecting the provider does not automatically switch every technician to SSO.

Administrators must change the authentication method for individual technicians through Workspace → Team.

Once a technician uses SSO, MFA enforcement is handled by the identity provider rather than Level's native 2FA settings.

Level also recommends maintaining a protected native-authentication administrator account for emergency recovery if the identity provider becomes unavailable.

Administrators can configure other core security controls under Settings → Security, including IP restrictions, remote control approval, and application inactivity timeouts.

Granular Technician and Group Permissions

Level's workspace permissions allow administrators to define roles with separate management, device, and group access permissions.

For example, an MSP can configure a technician role with access to assigned customer workstations while restricting broader administration, automation management, and other privileged functions.

Remote control and background management are separate device permissions.

Script access can also be controlled through script groups, allowing organizations to distinguish between technicians who can run approved scripts and those who can manage the script library.

CLI access is governed through a dedicated role permission. API keys are managed separately, with read-only and read-and-write access levels.

Administrator accounts bypass ordinary role-based restrictions, making careful assignment of administrator status essential.

These capabilities help MSPs reduce unnecessary privileges while retaining the flexibility required for different support responsibilities.

Encrypted Browser-Based Remote Control

Level's remote control allows technicians to connect to supported Windows, macOS, and Linux devices through a browser.

Connections use peer-to-peer communication where possible, with an end-to-end encrypted relay when direct connectivity is unavailable.

Because the remote control connection uses the Level agent, technicians do not need to configure public inbound RDP access for this workflow.

Configurable Remote Control Approval

Level supports notification-based and approval-based remote connections.

Administrators can configure the organization's default approval behavior and apply specific settings to device groups.

For environments requiring affirmative consent, Level provides a setting that blocks interactive remote control unless the end user approves it.

However, this setting applies to screen-sharing sessions, not background management tools.

MSPs should therefore combine remote control approval with correctly scoped permissions for terminal access, file management, and other background operations.

Activity Logging and Session Visibility

Level records remote session activity and relevant administrative actions, including script execution and configuration changes.

Its Activity system also provides before-and-after information for supported configuration modifications and code differences for saved script changes.

Completed remote control and terminal session records include session metadata, but not terminal command text or output.

These records support technician accountability and investigation workflows, although they should be combined with other security telemetry when comprehensive monitoring or command-level evidence is required.

Level's published security information also states that it does not record the screen contents of remote control sessions.

Technicians can review their own authenticated browser and CLI sessions under Settings → Sessions.

These session-management capabilities should be understood alongside the organization's broader identity and incident response processes.

Endpoint Management and Security Visibility

Level's security capabilities provide visibility into endpoint-related controls such as operating system patching, antivirus status, firewall state, and disk encryption.

Its endpoint management and patching capabilities can help MSPs address weaknesses that affect the security of managed devices.

These capabilities complement, rather than replace, a broader security program covering identity protection, endpoint detection, incident response, and customer-specific requirements.

For a detailed product-specific implementation walkthrough, see Level's Securing Level documentation.

MSPs looking to simplify remote support while maintaining clear administrative boundaries can explore Level's remote control capabilities and evaluate how they fit their existing security policies.

Final Thoughts

Secure remote access for MSPs depends on several controls working together: strong authentication, least-privilege permissions, protected technician workstations, encrypted connections, endpoint hardening, and reliable activity monitoring.

The most important distinction is between having security settings enabled and confirming that they actually prevent unauthorized access.

MSPs should also recognize that interactive remote control, background management, privileged automation, and programmatic access may require different authorization mechanisms.

Begin with the highest-impact controls, validate their enforcement, document exceptions, and review access as customer environments and technician responsibilities change.

A well-configured remote access environment allows technicians to provide efficient support without unnecessarily expanding the organization's security exposure.

Frequently Asked Questions

1. What is the most important security control for MSP remote access?

Mandatory multifactor authentication is one of the most important baseline controls because it reduces reliance on passwords alone. However, secure remote access also requires least-privilege permissions, protected technician devices, appropriate network restrictions, and monitoring. No single control is sufficient on its own.

2. Should MSPs use a VPN for every remote support session?

Not necessarily. The appropriate connection architecture depends on the remote access technology and customer requirements. Some agent-based remote support platforms provide encrypted connections without requiring a separate VPN or public inbound firewall rule. Administrative access still needs strong identity verification, authorization, and endpoint security.

3. How often should MSPs review remote access permissions?

MSPs should review permissions whenever technicians join, leave, or change responsibilities, and after significant customer or configuration changes. Periodic reviews should also examine privileged access, unused accounts, inherited permissions, script execution privileges, CLI access, API credentials, and authentication exceptions. The appropriate review interval depends on customer risk and contractual or regulatory requirements.

4. What should an MSP do if a technician account is compromised?

The MSP should promptly disable the affected account, revoke available sessions and credentials, and investigate recent access activity. It should assess whether any customer environments were affected, preserve relevant evidence, and follow established incident response and notification procedures. Because existing sessions, API credentials, and endpoint-level access may have separate lifetimes, containment should verify that unauthorized access has actually ended.