Skip to content

Latest commit

 

History

History
149 lines (120 loc) · 9.97 KB

File metadata and controls

149 lines (120 loc) · 9.97 KB

SOC2 (System and Organization Controls 2) Compliance Knowledge

Framework Overview

SOC2 is an auditing standard developed by the AICPA (American Institute of CPAs) for service organizations that store, process, or transmit customer data. Based on the Trust Service Criteria (TSC), published in 2017 and updated 2022.

Type I vs Type II:

  • Type I: Point-in-time assessment — controls are suitably designed as of a specific date
  • Type II: Period assessment (typically 6–12 months) — controls operate effectively over time
  • Relevance: Architecture-level CALM analysis maps to Type II control design objectives

Applies to: SaaS providers, cloud platforms, data processors, managed service providers — any service organization handling customer data.

Trust Service Categories

Common Criteria (CC) — Security (Always Required)

CC1 — Control Environment (CC1.1–CC1.5)

Organizational governance, board oversight, ethics, competence, accountability.

  • CALM observable: security-policy controls, governance metadata — CC1 is primarily organizational, low CALM signal

CC2 — Communication (CC2.1–CC2.3)

Internal/external communication of security objectives and responsibilities.

  • CALM observable: node metadata documenting security ownership — low CALM signal

CC3 — Risk Assessment (CC3.1–CC3.4)

Risk identification, fraud risk, change risk assessment processes.

  • CALM observable: risk-management controls on nodes — moderate CALM signal

CC4 — Monitoring (CC4.1–CC4.2)

Ongoing monitoring activities and evaluation of control deficiencies.

  • CALM observable: intrusion-detection, audit-logging controls — moderate CALM signal

CC5 — Control Activities (CC5.1–CC5.3)

Policies, procedures, and technology controls to mitigate risks.

  • CALM observable: Most controls map here as general — moderate CALM signal

CC6 — Logical and Physical Access Controls (CC6.1–CC6.8) — HIGH CALM SIGNAL

Criterion Description CALM Field
CC6.1 Logical access security software, infrastructure, architectures node.controls.access-control
CC6.2 New internal and external users registered and authorized node.controls.access-control
CC6.3 Role-based access and least privilege node.controls.access-control
CC6.4 Credentials and access revoked upon termination node.controls.access-control
CC6.5 Logical access restricted using multi-factor authentication node.controls.multi-factor-authentication
CC6.6 Logical access security measures protect against threats outside system boundaries node.controls.network-segmentation, deployed-in
CC6.7 Transmission, movement, and removal of information restricted connects[].protocol (HTTPS, TLS, mTLS)
CC6.8 Logical access security measures include data loss prevention controls node.controls.data-encryption

CC7 — System Operations (CC7.1–CC7.5) — HIGH CALM SIGNAL

Criterion Description CALM Field
CC7.1 Infrastructure components monitored to detect anomalies node.controls.intrusion-detection
CC7.2 Malicious software detected and addressed node.controls.malware-protection
CC7.3 Security event monitoring and analysis node.controls.audit-logging (SIEM)
CC7.4 Security incidents identified and reported node.controls.incident-response
CC7.5 Incidents communicated to affected parties node.controls.incident-response

CC8 — Change Management (CC8.1)

Changes managed to meet commitments; assessments before/after changes.

  • CALM observable: configuration-management, change-management controls — low-moderate CALM signal

CC9 — Risk Mitigation (CC9.1–CC9.2)

Risk mitigation activities, business disruption management, vendor/partner risk.

  • CALM observable: third-party-management controls, backup-recovery — moderate CALM signal

Availability (A) — Optional Category

Criterion Description CALM Field
A1.1 Current processing capacity and usage monitored node.controls.backup-recovery, node metadata
A1.2 Environmental threats monitored node.controls.backup-recovery
A1.3 Recovery plan procedures established and tested node.controls.backup-recovery, disaster-recovery

Confidentiality (C) — Optional Category

Criterion Description CALM Field
C1.1 Confidential information identified and protected during collection node.controls.data-encryption
C1.2 Confidential information disposed of at end of retention node.controls (data-retention)

Processing Integrity (PI) — Optional Category

Criterion Description CALM Field
PI1.1 Processing integrity policies established node metadata
PI1.2 System inputs validated node.controls (input-validation)
PI1.3 System processing complete, accurate, timely node metadata
PI1.4 System outputs validated node metadata
PI1.5 Stored data complete, accurate, current node.controls (data-integrity)

Privacy (P) — Optional Category (Rarely CALM-Observable)

P1.1–P1.2 cover privacy notice and consent. Organizational in nature; minimal CALM signal.

CALM Field Mapping Summary

SOC2 Criterion CALM Mapping Control Key
CC6.1 node.controls access-control
CC6.2 node.controls access-control
CC6.5 node.controls multi-factor-authentication
CC6.6 deployed-in relationships, network nodes network-segmentation
CC6.7 connects[].protocol (must be HTTPS, TLS, mTLS) protocol field
CC6.8 node.controls on database/storage data-encryption
CC7.1 node.controls on network/system nodes intrusion-detection
CC7.2 node.controls on service/system nodes malware-protection
CC7.3 node.controls audit-logging
CC7.4 node.controls incident-response
A1.1 node.controls on service nodes backup-recovery
A1.3 node.controls on service/database nodes disaster-recovery
C1.1 node.controls on database/data-asset nodes data-encryption

Closed Control ID Reference

Use ONLY the IDs in this table when citing SOC2 controls. Do not invent or extrapolate IDs beyond this list. CITE EXACTLY AS SHOWN in the "Citation Format" column.

Control ID Full Name CALM Field Citation Format
CC6.1 Logical access security software, infrastructure, architectures node.controls.access-control "SOC2 TSC CC6.1"
CC6.2 New internal and external users registered and authorized node.controls.access-control "SOC2 TSC CC6.2"
CC6.3 Role-based access and least privilege node.controls.access-control "SOC2 TSC CC6.3"
CC6.4 Credentials and access revoked upon termination node.controls.access-control "SOC2 TSC CC6.4"
CC6.5 Logical access restricted using multi-factor authentication node.controls.multi-factor-authentication "SOC2 TSC CC6.5"
CC6.6 Logical access security measures protect against external threats node.controls.network-segmentation "SOC2 TSC CC6.6"
CC6.7 Transmission, movement, and removal of information restricted connects[].protocol "SOC2 TSC CC6.7"
CC6.8 Data loss prevention controls for logical access node.controls.data-encryption "SOC2 TSC CC6.8"
CC7.1 Infrastructure components monitored to detect anomalies node.controls.intrusion-detection "SOC2 TSC CC7.1"
CC7.2 Malicious software detected and addressed node.controls.malware-protection "SOC2 TSC CC7.2"
CC7.3 Security event monitoring and analysis node.controls.audit-logging "SOC2 TSC CC7.3"
CC7.4 Security incidents identified and reported node.controls.incident-response "SOC2 TSC CC7.4"
CC7.5 Incidents communicated to affected parties node.controls.incident-response "SOC2 TSC CC7.5"
CC8.1 Changes managed to meet commitments node.controls.change-management "SOC2 TSC CC8.1"
CC9.1 Risk mitigation activities node.controls.risk-management "SOC2 TSC CC9.1"
CC9.2 Business disruption and vendor/partner risk node.controls.third-party-management "SOC2 TSC CC9.2"
A1.1 Current processing capacity and usage monitored node.controls.backup-recovery "SOC2 TSC A1.1"
A1.2 Environmental threats monitored node.controls.backup-recovery "SOC2 TSC A1.2"
A1.3 Recovery plan procedures established and tested node.controls.disaster-recovery "SOC2 TSC A1.3"
C1.1 Confidential information identified and protected node.controls.data-encryption "SOC2 TSC C1.1"
C1.2 Confidential information disposed at end of retention node.controls (data-retention) "SOC2 TSC C1.2"

Notes for LLM Analysis

  • Focus on CC6 and CC7 — These are the highest CALM-signal criteria. CC6 maps directly to access controls, MFA, and protocol security. CC7 maps to monitoring, malware, and incident response.
  • CC1–CC5 are organizational controls — Policies, governance, risk programs. CALM architecture files rarely contain enough signal to assess these; note as "not assessable from architecture" rather than flagging as gaps.
  • CC6.7 is the protocol criterion — Any connects relationship using HTTP, FTP, or TCP (unencrypted) fails CC6.7. HTTPS, TLS, mTLS, and SFTP are compliant.
  • Availability category (A) is optional but commonly required by B2B SaaS customers. Assess if backup-recovery or disaster-recovery controls are present.
  • Confidentiality (C) is optional but relevant to any architecture handling customer data. data-encryption on database nodes is the primary signal.
  • Do not assess Privacy (P) from architecture — P criteria are consent/notice based, not architecture-observable.
  • Type II relevance — CALM can only assess control design (Type I evidence). Note this limitation in findings.