CM0024 Anti-counterfeit Hardware Counterfeit electronic components represent a direct supply chain threat to space mission integrity, introducing hardware that may fail prematurely, perform outside specification, or contain malicious functionality deliberately embedded by an adversary during manufacture or distribution. A formal anti-counterfeit program must establish policy and procedures that span the entire component acquisition and integration lifecycle, from supplier qualification and procurement through incoming inspection, storage, and installation. The program must address two distinct but related risks: counterfeit components that fail to perform their intended function, degrading mission reliability; and deliberately tampered components that introduce malicious hardware functionality or create pathways for malicious code execution. Anti-counterfeit controls must include measures appropriate to component criticality and supply chain risk to authenticate components, detect evidence of tampering, and resist unauthorized modification. Detection and prevention must be treated as complementary objectives: prevention through qualified sourcing and procurement controls, detection through inspection and authentication techniques applied before components enter the system.
CM0026 Original Component Manufacturer Hardware components that cannot be sourced directly from the original component manufacturer (OCM) or an authorized franchised distributor, and software that cannot be obtained from the original publisher, developer, or an authorized distribution channel, represent elevated supply-chain risk and must not be procured or incorporated into the mission system without documented approval from the program’s supply-chain governance authority. Sourcing hardware from the OCM or authorized franchised distributors provides greater assurance of component authenticity, traceability, and conformance to specification. Obtaining software from the original publisher, developer, or an authorized distribution channel similarly reduces the risk of unauthorized, altered, fraudulent, or malicious software. Deviations from these approved sourcing channels introduce additional supply-chain risk that must be assessed before acceptance. The approval process for non-OCM-sourced items must evaluate the specific risk posed by the alternative source, the criticality of the component or software to mission function, the availability and adequacy of compensating inspection and authentication measures, and whether a compliant source can be identified before accepting the deviation. This governance requirement applies to hardware components, firmware, and software, although the applicable sourcing and authentication methods differ. Hardware controls should address component authenticity and traceability, while software controls should address publisher or developer provenance, distribution-channel integrity, license legitimacy, and cryptographic verification where available.
CM0027 ASIC/FPGA Manufacturing Custom application-specific integrated circuits (ASICs) should be fabricated through accredited trusted foundries, and field-programmable gate array (FPGA) devices should be procured through trusted suppliers with documented fabrication provenance, to reduce the risk of hardware Trojan insertion or unauthorized modification. Unlike software, hardware trojans embedded during semiconductor manufacturing are extremely difficult to detect through functional testing alone, as they may be designed to activate only under specific operational conditions or remain dormant indefinitely; the integrity of the fabrication source is therefore a primary defense. Trusted foundry accreditation provides assurance that the accredited fabrication activities are subject to security controls intended to reduce the risk of unauthorized modification. Assurance for design, intellectual property, aggregation, packaging, assembly, testing, and distribution must be addressed through trusted suppliers or other controls applicable to those lifecycle stages. This requirement applies to custom ASICs and to the base silicon used in FPGA implementations. The programmable design loaded onto an FPGA requires separate protection because trusted fabrication of the device does not establish the integrity or authenticity of the configured bitstream.
CM0028 Tamper Protection Tamper protection encompasses physical and logical controls that detect, deter, and respond to unauthorized modification of mission hardware and software throughout the acquisition, transit, storage, integration, and operational phases of the mission lifecycle. Physical tamper protection requires inspection of hardware at defined custody transfer points to detect evidence of unauthorized access or modification, and requires the use of tamper-evident packaging and sealing mechanisms during shipping and receiving to ensure that any interference with equipment in transit is detectable upon arrival. Logical tamper protection addresses the integrity of software and firmware through layered techniques including code obfuscation, integrity checks that verify software has not been altered from its approved state, and runtime integrity monitoring mechanisms such as self-checking code. Watchdog processes may supplement these controls by detecting abnormal execution or failure to complete expected processing, but do not by themselves verify software integrity. Physical and logical tamper protections are complementary and should be applied together where warranted by system criticality, threat exposure, and lifecycle conditions. Physical access to hardware can enable logical tampering, while failure to verify software integrity can reduce the assurance provided by physical controls.
CM0011 Vulnerability Scanning Vulnerability scanning systematically identifies known security weaknesses in commercial off-the-shelf (COTS) and open-source software (OSS) components, including vulnerable dependencies and outdated software versions, across spacecraft and ground system environments. Custom-developed code requires separate analysis approaches, such as static and dynamic code analysis, that fall outside the scope of this countermeasure; however, COTS, OSS, and third-party dependencies incorporated into custom-developed software remain within the scope of vulnerability scanning and SCA. Scanning programs should incorporate software composition analysis (SCA) to detect vulnerabilities introduced through third-party libraries and dependency chains, which represent a significant and frequently underestimated attack surface in modern space systems. Scanning tools and processes should conform to recognized interoperability standards that support enumeration of platforms, software flaws, and configuration weaknesses; standardized formatting of checklists and test procedures; and consistent measurement of vulnerability impact. Adherence to these standards enables automation of key vulnerability management workflow steps, facilitates tool interoperability across the mission ecosystem, and produces outputs that are comparable across programs and organizations.
Vulnerability scanning shall be performed against the approved as-built or deployed software baseline using methods that do not jeopardize mission operations. Active scanning of live spacecraft, safety-critical ground systems, or operational technology should occur only after the scanning method has been authorized and evaluated for operational impact. Where live scanning presents unacceptable risk, scanning should be performed against representative software images, firmware packages, or test environments that match the approved operational baseline.
CM0015 Software Source Control Binary or machine-executable code obtained from sources that provide no warranty and no access to the corresponding source code must not be incorporated into spacecraft or ground systems. This prohibition addresses a fundamental software assurance gap: without source code, missions cannot perform source-based static analysis or independently review and modify implementation details, and may have reduced ability to assess, repair, or extend the software. Code from sources with limited or no warranty and no source code provision may be difficult to independently analyze, repair, or extend and leaves the mission dependent on supplier assurances and remediation capabilities. This countermeasure applies throughout the software supply chain, including components integrated by subcontractors.
CM0018 Dynamic Testing Dynamic analysis subjects executing software and firmware to active testing conditions to identify weaknesses and vulnerabilities that static analysis alone cannot detect, including runtime memory corruption, logic errors, timing vulnerabilities, and interface behaviors that only manifest under operational conditions. Applicable techniques include simulation-based testing, penetration testing, fuzz testing, and adversarial emulation using realistic threat actor tactics, techniques, and procedures (TTPs). Dynamic testing must be applied to all software and firmware categories: mission-developed code, open-source components, commercial off-the-shelf (COTS) software, and third-party developed code. Dynamic testing should be performed on potential system elements before acceptance, include realistic simulation of known adversary TTPs, and tools, and continue throughout the lifecycle on applicable physical and logical systems, elements, and processes. Full-scale hardware integration testbeds, commonly referred to as flat satellite (FLATSAT) environments, provide high-fidelity physical test platforms for dynamic analysis; digital twins implemented via instruction set simulation, also known as emulation, provide a flexible and scalable alternative environment capable of supporting TTP execution across a broad range of test scenarios without requiring physical hardware availability.
CM0019 Static Analysis Static source code analysis examines software without executing it, identifying security-relevant weaknesses in the codebase before they can be exploited at runtime. Static analysis must be performed across all available source code. The mission’s weakness prioritization (i.e., CM0016) process should guide tool configuration and finding triage without unnecessarily excluding other weakness classes supported by the tools. The static analysis program must employ no fewer than two distinct tools with complementary language and weakness-detection capabilities. The tools should be applied at defined points throughout the development and maintenance lifecycle to provide overlapping and complementary analysis coverage and reduce the likelihood that exploitable weaknesses persist into integration or deployment. Tool selection should account for language compatibility, weakness class coverage, and the ability to produce findings in formats that support triage and remediation tracking. Static analysis should be integrated into the software build pipeline as an automated, recurring activity rather than conducted as a periodic manual exercise, ensuring that new code contributions are analyzed continuously throughout development.
CM0021 Software Digital Signature Each software or firmware image subject to installation or update control shall be verified using an approved digital signature and an approved trust anchor before installation or activation. The trust anchor may be represented by a certificate or by a directly provisioned verification key, depending on the approved trust architecture. Successful verification establishes that the image was signed by an authorized signing identity and has not been modified since signing; it does not establish that the signed code is non-malicious, vulnerability-free, or operationally safe. Signature verification confirms the integrity and approved origin of the software but does not, by itself, prevent installation of an older validly signed version; update authorization and rollback protections must be enforced separately.
The verification mechanism must be implemented such that it cannot be bypassed through operational commands, configuration changes, or software updates, and must reject any component whose signature is absent or invalid, or whose signing certificate or trust anchor is not recognized and approved by the mission. Digital signature enforcement complements but is distinct from the boot-time chain of trust established through secure boot; it applies to software installation and update events throughout the operational lifecycle, not only at system startup.
CM0046 Long Duration Testing Long duration testing subjects spacecraft software, firmware, hardware, and relevant integrated ground interfaces, or representative simulation and emulation environments, to extended test execution of 30 days or more to expose security and reliability defects that may manifest only after prolonged operation or specific time-dependent conditions. Race conditions, memory or resource leaks, time-dependent state corruption, resource exhaustion, counter rollover, and time-triggered malicious behavior may not manifest during short-duration testing because their activation depends on accumulated runtime, rare timing interactions, or gradual changes in system state. Long duration testing increases the opportunity to expose these conditions before deployment and complements static analysis, formal analysis, stress testing, fault injection, and targeted rollover testing. Testing should use the highest-fidelity environment appropriate to the test objectives. Flight-representative hardware should be used where hardware timing, device behavior, or integration effects are material; validated simulation or emulation may be used for conditions that it represents with sufficient fidelity. Differences between the test environment and operational system must be documented and considered when interpreting results.
CM0032 On-board Intrusion Detection & Prevention An on-board intrusion detection and prevention system (IDS/IPS) monitors mission-critical spacecraft components and systems, generates and stores audit records, and supports mission-approved responses to detected threats. Depending on the mission architecture, threat, and availability of ground support, responses may be autonomous, ground-directed, or a combination of both. The system should address both known attack patterns and previously unseen anomalous behavior through complementary signature-based and behavior- or anomaly-based detection methods. Machine learning or adaptive technologies may be used when their performance, resource consumption, and failure behavior have been validated for the mission environment. Detection and response coverage should address applicable adversary activities across the attack lifecycle, including initial access, execution, persistence, defense evasion, and exfiltration. The on-board IDS/IPS must be integrated with the spacecraft's traditional fault management system to provide a unified approach to anomaly response, ensuring that cyber-triggered responses are compatible with fault management logic and do not produce unintended effects or fratricide against the spacecraft's own systems; countermeasures that are incompatible with fault management are considered unsafe and must not be executed autonomously. The response hierarchy must prioritize vehicle safety and continued mission operations. Advanced containment or deception responses may be considered when they can be executed without unacceptable mission risk. The system should preserve evidence that supports post-event analysis, threat characterization, and potential attribution by authorized ground support.
CM0068 Reinforcement Learning A reinforcement learning (RL) agent deployed within the spacecraft or ground system can provide an adaptive, autonomous anomaly detection and response capability that identifies anomalous events, including malicious data inputs and injected commands, and redirects affected processes to proceed safely by ignoring or isolating the malicious input. An RL agent learns a response policy that maps observations to actions according to its training environment and reward function. It may generalize to scenarios not explicitly included in training, but its ability to detect or respond correctly to novel attacks or conditions outside the validated operational envelope must not be assumed. Anomaly detection may be incorporated into the RL architecture or provided by a separate monitoring function. Effective deployment requires separate protections against compromise of the training process and manipulation of observations presented to the deployed agent. Online learning or policy adaptation should be disabled unless specifically authorized, bounded, and validated. Agent-selected responses must be constrained by a trusted safety mechanism.