Every CSIRT has its own distinct composition. However, most CSIRTS combine people and policies in ways that clearly define their missions. In terms of people, a CSIRT usually has a core group of dedicated members that is supplemented by experts who work with the CSIRT on an as-needed basis. The team members invariably come from different backgrounds and skill sets. For instance, some are experts in defending Windows systems, while others know about Linux. The core team may be assigned to the CSIRT full-time, but that occurs mostly at very large organizations. In most cases, CSIRT team members have “day jobs” in IT and cybersecurity departments.
Regarding policies, in addition to the CSIRT’s mission statement and written definitions of relevant constituencies, the CSIRT will create and maintain a set of documents that establish how the CSIRT functions. For example, a CSIRT usually has a central incident response plan that declares, in writing, how the team handles on-site incident response processes versus handling incident response by phone. The plan reveals how the CSIRT coordinates incident response by allocating team resources across multiple constituent groups.
As part of this effort, a CSIRT will publish forms and contact directories to share with each group it services in the organization. This may sound like an unnecessary step, but experience has shown that key stakeholders often don’t know whom they are supposed to contact if there is a security incident. For example, if the CEO’s laptop is stolen, will their administrative assistant know to contact the CSIRT, or will he simply call the IT help desk — who themselves may need clear prompts to activate the CSIRT.
A CSIRT also compiles many internal policy and procedure documents. These cover how the CSIRT functions, how it responds to incidents, who does what, how to prepare incident reports, and more.
The specifics of these policies will vary according to the CSIRT structure. Common organizational approaches to establishing a CSIRT include:
- Centralized CSIRT — a single team is used for the whole organization.
- Distributed CSIRT — multiple teams are based in different regions and/or assigned to different business units.
- Coordinating CSIRT — one central CSIRT coordinates the work of subordinate CSIRTs.
- Hybrid CSIRT — in this combination of centralized and decentralized models, the central CSIRT is full-time and distributed CSIRTs are on call.
- CSIRT/SOC hybrid — the SOC is on task for day-to-day incident responses and alert management, but the CSIRT gets activated when certain serious conditions are met.
- Outsourced CSIRT — an external provider performs the work of the CSIRT, an approach that may be suitable for a smaller organization. Outsourced CSIRTs can also perform tasks that fall outside an internal CSIRT’s skill set, e.g., ransomware remediation or digital forensics.