Industry
MSPs & MSSPs
MSPs and MSSPs administer systems that can reach many customers, making their own control environment a critical security boundary. Protection should focus on remote-management infrastructure, privileged credentials, tenant separation, and clearly assigned response authority rather than assuming centralized administration is inherently safe.
Field notes · 01
Treat management tooling as production-critical infrastructure
Inventory remote monitoring and management systems, professional services automation, deployment tools, support portals, credential stores, and their integrations. Determine which identities can execute commands, distribute software, or change policy across clients. Restrict those permissions and monitor the activity that uses them.
Librarian can help organize management dependencies, and Argus can inform exposed-secret investigations. Keep administration separate from ordinary browsing and email. A compromised support account should not automatically provide unrestricted access to every customer, even when centralized tooling makes that access convenient.
- Use individual privileged accounts and strong authentication.
- Review deployment scripts, API tokens, and third-party management integrations.
Field notes · 02
Demonstrate tenant separation in daily operations
Test isolation across access permissions, logs, tickets, reports, backups, and automation jobs. Tenant separation is not only a dashboard feature: engineers must avoid copying secrets or customer evidence into the wrong workspace. Define how emergency access is requested, approved, recorded, and removed.
Use Relay with least-privilege integrations and clearly documented customer authorization. Sleight or Verification & Response activities must be explicitly scoped to the customer environment involved. Do not infer that a service agreement authorizes every type of testing, data collection, or automated remediation.
- Separate credentials and recovery access between customers.
- Review technician access when assignments and customer contracts change.
Field notes · 03
Agree response authority before a multi-client incident
Document which actions the provider may take, which require customer approval, and how emergency decisions are escalated. Prepare for a compromise of the provider's own tooling, including the possibility that normal communications and ticketing cannot be trusted. Maintain verified customer contacts independently.
Dewey can support coordination and Growler can assist with evidence handling, but human authorities retain consequential decisions. Judge can help organize posture evidence without guaranteeing contractual compliance or insurance outcomes. Confirm deployment-specific multi-tenant, white-label, or OEM arrangements rather than treating them as universal capabilities.
- Exercise revocation of shared management access.
- Test restoration and customer notification workflows without mixing tenant information.
Further reading
These links are background sector guidance, not validation of every recommendation or Aktoh capability. Aktoh role descriptions outline possible workflows; confirm supported connections, access, and scope before relying on them. The sources do not endorse Aktoh. Requirements depend on your organization, contracts, and jurisdiction; this page is not legal advice.
What we see here
Use the sector guidance above to identify critical systems, access risks, and recovery priorities. Coverage depends on the systems you connect and the workflows you configure.
How the fleet responds
Agents investigate and prepare recommendations in parallel. System-changing actions require approval by default; you choose which action classes can run automatically.
What you get back
Review plain-language findings and available audit records. Relevant agents can help organize evidence; your team remains responsible for compliance decisions.
Why there are no logos on this page
We do not identify customers without permission. Security scenarios on this site are illustrative unless accompanied by verified supporting evidence; they are not proof of customer outcomes or performance guarantees.
Start free. Add credits only when the work justifies it.
Run the whole fleet across unlimited nodes at zero platform fee — you pay only for tokens. Move to a pooled plan when the usage justifies the lower rate.