Topic · ssdf

NIST SP 800-218 (SSDF)

Secure Software Development Framework — federal procurement attestations and supplier requirements driven by EO 14028.

RSS feed ↗

trade-press
DOD CIO binds developers to full liability for AI-generated code
The instruction treats every line of AI-suggested code as unverified input and puts the developer on the hook for vulnerabilities, bugs, and IP contamination, no carve-out for the model having written it.
Pentagon CIO Kirsten Davies issued binding instruction on Accelerated Mission Software, effective Sept. 8, requiring that all AI-generated or AI-modified code undergo the same security review, human gate-check, and continuous monitoring as manually written code. Developers and teams "remain fully accountable" for AI output; sensitive DOD code and schematics can't touch unapproved commercial AI tools. The directive mandates an "AI evidence package" (model versions, datasets, prompts) analogous to SBOMs, establishing a transparency baseline that goes beyond what most industry shops do today.
SEP 14
trade-press
NIST Builds SBOM Use Cases for DevSecOps Pipelines
The NCCoE's example builds move beyond generating an SBOM to integrating cryptographic signing at each build stage and operationalizing vulnerability detection, bridging the June cyber EO's SSDF tasking to actual infrastructure.
NIST's National Cybersecurity Center of Excellence is publishing example builds that show how to use SBOMs for active supply-chain scanning and vulnerability management, rather than simply generating and storing them. The builds integrate cryptographic hashing and signing at each stage of the DevSecOps pipeline. The work responds to the Trump administration's June 6 cyber executive order directing NCCoE to demonstrate SSDF implementation practices. Michael Ogata, NCCoE computer scientist, described the efforts at the Billington Cybersecurity Summit, noting that two example builds have been published so far under the DevSecOps project, with more planned for vulnerability management workflows.
SEP 9
trade-press
The quantum attack surface is bigger than it seems
EO 14412 and M-26-15 prioritize high-value system migration, but the trust infrastructure that keeps federal networks authenticating, routing, and verifying software updates sits outside the mandate, and adversaries won't wait for separate guidance.
The White House and OMB have set hard deadlines for migrating high-value and high-impact federal systems to post-quantum cryptography, key establishment by 2030, digital signatures by 2031. But the mandates are silent on the broader trust infrastructure those systems depend on: PKI, device identity, code-signing, network routing, and forensic logs. A cryptographically relevant quantum computer wouldn't need to crack the prioritized systems directly. It could forge certificates, redirect traffic, or push malware through trusted update channels, and the agencies that followed the mandate to the letter would still be exposed. With Google, Microsoft, and Cloudflare targeting 2029 for their own PQC migrations, the gap between what's mandated and what's actually protected is a vulnerability window measured in years.
SEP 1
trade-press
Army all but abandons custom-built software for commercial buys
Driscoll's policy update makes commercial solutions the default acquisition path and tells cybersecurity officials to clear products faster, but the continuous ATO framework that would make that work is still in pilot.
Army Secretary Dan Driscoll's July 22 policy directive, published last week, orders the service to default to commercial solutions openings and other transaction authorities for software procurement, reserving custom development for cases where no commercial product meets operational needs. The memo also instructs cybersecurity officials to lean on continuous ATOs and reciprocity agreements to accelerate deployment. But the Army's continuous ATO framework (launched in pilot under CIO Leonel Garciga) hasn't yet reached production, leaving the security review model the policy depends on in an unfinished state.
AUG 27
regulator
CISA Lays Vulnerability Baseline Before AI Floods the Pipeline
The review frames Secure by Design not as an aspirational slogan but as the only systemic alternative to a reactive patching model that won't scale.
CISA published a vulnerability review analyzing FY2024, 2025 data to establish a baseline of today's vulnerability landscape before AI-enabled discovery becomes more widespread. The review identifies common software weaknesses and shows organizations how to prioritize remediation using BOD 26-04's four criteria: exposure status, KEV catalog status, automation potential, and technical impact. The framing matters: CISA is telling organizations to fix classes of vulnerabilities systemically, not play whack-a-mole with individual CVEs.
AUG 26
trade-press
FedRAMP 20x ends compliance theater with 2, 4 day vuln deadlines
If your security team reviews findings weeks after engineering ships code, your FedRAMP authorization is at risk, the pipeline is the control now.
FedRAMP Director Pete Waterman told vendors at Carahsoft's FedRAMP Summit that those who can't fix critical, internet-facing vulnerabilities within days don't belong in the federal marketplace. FedRAMP 20x codifies that expectation: risk reduction must begin within two to four days, and system security state must be verified at least every three days. The implication is structural, vendors whose compliance staff sit apart from engineering cannot sustain that pace. Continuous authorization requires continuous delivery, and the deployment pipeline itself becomes the control.
AUG 25
trade-press
AI Coding Tools Widen DoD's Container Vulnerability Gap
AI tools that pull from unhardened open-source images are scaling vulnerability backlogs faster than DoD's patching cadence can absorb, and the fix isn't new policy, it's SSDF fundamentals that AI supply chains make harder to enforce.
AI coding tools default to unhardened open-source container images, pulling in CVEs faster than traditional vulnerability management can handle. The CVE list averaged 132 new entries per day in 2025, triple the 2020 figure, and AI tooling's daily-to-weekly update cycle compounds the gap. For DoD authorizing officials, fewer AI tools clear ATO as container stacks fail compliance at mission tempo. The fix, hardened curated images and contextual CVE evaluation, is SSDF fundamentals that AI supply chains have made harder to enforce.
AUG 25
trade-press
CISA gives OSS equal footing with proprietary code
The guidance tells agencies to stop treating open source as an exotic risk category, the same supply-chain principles apply, and OSS has the advantage that you can actually read the code.
CISA published "Open Source Software: Security Principles and Practices" on July 30, giving federal agencies a lifecycle framework for using, evaluating, contributing to, and publishing open source software. The guidance aligns with Biden's EO 14144 and Trump's June 2025 EO 14306. The headline move: CISA explicitly states OSS carries no more or less risk than proprietary software, the distinction is that with OSS, agencies can directly assess code quality instead of relying on vendor assurances. That's a quiet but significant policy signal after years of agencies treating OSS as inherently suspect.
JUL 31
vendor
AWS prescribes 24-hour cooldown for npm and pip packages
Amazon Linux's advice targets direct installs only, transitive dependencies pulled by pinned packages get no such guard, and the post is silent on whether Inspector's malicious-package detections make the window redundant.
AWS published a Security Blog post recommending that npm and pip users configure a 24-hour dependency cooldown, telling the package manager to skip versions published in the last day. The rationale: recent supply-chain attacks on npm were all detected and removed within hours, so a one-day lag would have neutralized them. The post shows how to override the cooldown for security patches and acknowledges that the approach leaves transitive dependencies unguarded.
JUL 29
regulator
CISA Issues First SBOM Baseline Update Since 2021
The multi-agency update reflects five years of tooling maturity and stakeholder input but leaves open whether existing SBOMs must be rebuilt to the new floor.
CISA, the NSA, the FBI, and international partners released the 2026 Minimum Elements for a Software Bill of Materials, replacing the 2021 NTIA baseline that governed SBOM expectations for five years. The update incorporates feedback from a 2025 public comment period and reflects current SBOM tooling while preserving the original document's core principles. AI systems and SaaS products, the guidance notes, may require elements beyond the minimum. For software producers and federal suppliers, this is the new reference point for what a conforming SBOM contains. The guidance is silent on whether existing SBOMs must be retroactively updated.
JUL 29
trade-press
Sandworm_Mode targets AI coding assistants and CI/CD pipelines
The compliance problem is not AI novelty; it is secret sprawl moving faster than review, logging and provenance can explain.
CyberScoop reports that CrowdStrike reviewed Sandworm_Mode, a self-propagating worm found by Socket in February, that targets AI coding assistants, LLM API keys, cloud credentials and CI/CD systems. Developers, DevOps teams and software suppliers face the operational exposure. The ugly part is familiar: the malware hides inside normal automated dependency activity, and CrowdStrike has not pinned down its operator, intent or current activity.
JUL 22
standards
CIS, SAFECode update Secure by Design for AI
The available source describes a guide update, not a disclosed mandate, deadline or required AI tool.
CIS and SAFECode updated Secure by Design: A Developer’s Guide to Building Safer Software to address AI’s role in software security. The change affects readers who track secure software development guidance, including independent software vendors, primes and contractors. The source does not disclose specific tool requirements, implementation timelines or procurement consequences.
JUL 13
trade-press
AI-generated code forces CISOs to measure risk velocity
The useful warning is not that AI writes bad code; it is that AppSec queues were never sized for machine-speed change.
A CyberScoop op-ed argues CISOs should treat AI-generated code as a high-risk input, with automated testing, dependency checks, fast remediation and policy gates before production. The piece says the relevant metric is “risk velocity,” citing familiar failure modes such as weak validation, hard-coded secrets, vulnerable dependencies and package hallucinations. The claim is governance, not magic: faster code creation makes old review bottlenecks worse.
JUL 13
trade-press
Zabierek pushes state software liability for secure-by-design rules
The federal strategy keeps naming manufacturers, but the legal pressure may arrive first from state product and privacy law.
Former CISA secure-by-design co-lead Lauren Zabierek told Inside Cybersecurity that states should explore software liability rules, using privacy and product-liability frameworks to push manufacturers toward secure-by-default development. The argument lands on state procurement offices, legislatures and software vendors, especially around critical infrastructure. Her uncomfortable point is capacity: the Trump resiliency order puts more security and resilience on states she says do not yet have the resources.
JUL 9
trade-press
Former CISA official Cable presses DoD on secure-by-design AI coding guardrails
CISA’s old point follows the code generator: security becomes useful only when it constrains the first draft.
Former CISA senior official Jack Cable, now co-founder of Corridor Security, told Inside Cybersecurity that DoD should apply secure by design principles as it scales AI-assisted coding, including its Feb. 19 call for AI-enabled coding capabilities. The advice has a business model attached. It still lands cleanly on developers, acquisition teams and reviewers who must guide AI agents before generated code reaches defense software pipelines.
JUL 6
trade-press
NIST NCCoE advances Cyber AI Profile, agent identity work
NIST is treating AI as plumbing for cybersecurity programs, which means old control maps will need uncomfortable new seams.
Federal News Network reports that NIST’s National Cybersecurity Center of Excellence is moving six AI-cyber projects, including a summer revision of the Cyber AI Profile and agent identity work that drew comments from more than 600 organizations. Primes, contractors, managed service providers and C3PAOs should read this as early architecture guidance, not certification text. The unresolved question is whether NIST will map existing Cybersecurity Framework, DevSecOps and supply-chain practices tightly enough to change compliance evidence.
JUN 30
regulator
CISA turns Secure by Design toward EdTech pledges
The whitepaper still reads as guidance, but CISA is teaching buyers how to make voluntary baselines feel expensive to ignore.
CISA released the second iteration of its Secure by Design whitepaper, expanding co-endorsement to 18 U.S. and international partners and adding more detail on evidence artifacts. It also launched a K-12 EdTech pledge asking software manufacturers to commit to secure development practices, publish a roadmap, and stop charging extra for basic security features. Primes and ISVs should read this less as a pamphlet and more as early procurement pressure.
JUN 29
regulator
CISA sets IT SSGs for secure software development
CISA can call them voluntary while critical-infrastructure buyers convert the goals into supplier evidence requests.
The Cybersecurity and Infrastructure Security Agency (CISA) released voluntary information technology (IT) Sector Specific Goals (SSGs) on Jan. 7 for secure development, product design and software supply chain risk management. They reach primes, subcontractors, independent software vendors and contractors serving critical infrastructure, with practices covering development-environment separation, trust-relationship monitoring, phishing-resistant multi-factor authentication, secrets management and supply chain risk management. The next 18 months are a procurement story: customer contracts can mandate what CISA frames as voluntary.
JUN 29
trade-press
GAO urges CISA to resolve OMB-NIST SBOM conflict
Federal procurement moved before the compliance mechanics existed, leaving vendors to reconcile static paperwork with operational vulnerability data.
Inside Cybersecurity reports that the Government Accountability Office found 17 of 24 surveyed agencies saw a conflict between Office of Management and Budget (OMB) and National Institute of Standards and Technology (NIST) Software Bill of Materials (SBOM) guidance. OMB treated SBOMs as optional static artifacts; NIST treated them as compliance data for vulnerability management. Contractors, primes, subs and Certified Third-Party Assessment Organizations (C3PAOs) get the result: parallel agency collection workflows unless the Cybersecurity and Infrastructure Security Agency (CISA) gives agencies a common implementation model.
JUN 29
trade-press
Open-source attacks expose supply-chain accountability gaps
The hard part is assigning responsibility for code everybody uses, nobody owns, and procurement still treats as somebody else’s evidence problem.
CyberScoop reports that open-source software attacks are rising as experts question how much government can do without heavier vendor accountability. Former CISA open-source security staffers described years of underinvestment and setbacks under President Donald Trump, while Project Glasswing said it found 6,202 high- or critical-severity vulnerabilities across more than 1,000 projects. Federal buyers, software suppliers and maintainers all feel the risk; only one of those groups usually gets a contract clause.
JUN 24
vendor
EO 14409 sets 2030 contractor deadline for post-quantum FIPS
The advisory phase is over, and OMB now has to turn crypto inventory work into procurement evidence.
Executive Order 14409, signed June 22, 2026, gives agencies until Dec. 31, 2030, to move their most sensitive systems to post-quantum cryptography for encryption and until Dec. 31, 2031, for authentication. Federal contractors must meet post-quantum Federal Information Processing Standards by the end of 2030 or risk procurement ineligibility. The hard part is the evidence: the Office of Management and Budget still has to say how compliance gets verified.
JUN 23
trade-press
TeamPCP compromises more than 1,000 open-source packages
For federal suppliers, Secure Software Development Framework evidence starts to matter when engineering can prove what automated build systems consumed and published.
CyberScoop reports that TeamPCP has compromised and injected malicious code into more than 1,000 open-source packages in under four months, abusing continuous integration and continuous delivery (CI/CD) automation and artificial intelligence-assisted dependency habits. Developers, maintainers and software buyers that ingest packages at speed are affected. For Secure Software Development Framework (SSDF) programs, the gap is verification: a familiar package name does not prove a clean build path.
JUN 18
trade-press
Claude Code’s 30-plus patches expose AI agent update gap
For regulated teams, the risk is the week between a quiet fix and the next version they are allowed to run.
CyberScoop reports that Backslash Security reviewed Claude Code update logs and traced more than 30 security-relevant patches between April and early June 2026, including data poisoning, prompt injection, arbitrary code execution, OAuth leakage and a backslash bypass for destructive commands. Anthropic patched the issues, but did not separately publicize them. Teams that freeze AI coding-agent versions for vetting, air-gapped use or stability inherit the gap that auto-update users largely avoid.
JUN 16
vendor
Microsoft flags Claude Code Action secret-exposure flaw
Agentic CI/CD turns issue comments into an input boundary, and old secret-scrubbing assumptions do not cover that model.
Microsoft Threat Intelligence says Anthropic’s Claude Code GitHub Action could expose CI/CD workflow secrets when agents processed untrusted GitHub issues, pull request text, or comments. The gap was in the Read tool, which Microsoft says could access /proc/self/environ and read ANTHROPIC_API_KEY and possibly other runner credentials. Anthropic mitigated it in Claude Code 2.1.128 by blocking sensitive /proc files; teams using AI-assisted GitHub workflows need to separate untrusted content from secrets and tool access.
JUN 5
trade-press
House NDAA amendment targets CMMC cost burden
Congress is no longer just watching CMMC rollout mechanics; it is asking whether the market structure is pricing out small suppliers.
Inside Cybersecurity reports that House Armed Services advanced the fiscal 2027 National Defense Authorization Act, 44-12, with a Rep. Jeff Crank amendment requesting a Defense Department briefing on Cybersecurity Maturity Model Certification reforms by March 1, 2027. The amendment cites excessive regulatory burdens, small-business disadvantage, and fee transparency concerns involving the Cyber AB and CMMC Third-Party Assessment Organizations. Primes, subs, assessors, and the Cyber AB get the message: CMMC cost is now a Hill oversight target.
JUN 5