sota-api-design
State-of-the-art API design and audit guidance (2026) covering REST/HTTP, GraphQL, gRPC, WebSockets/SSE/realtime, webhooks, versioning/evolution, and API…
State-of-the-art security & compliance engineering (2026) for the cybersecurity control frameworks and product-security regulations that drive architecture, code, and CI gates — not the organizational policy binder. Use when work must satisfy or be audited against NIST CSF 2.0,
$ npx -y skills add martinholovsky/SOTA-skills --skill sota-security-compliance --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/sota-security-complianceContext preview
The summary Claude sees to decide when to auto-load this skill.
State-of-the-art security & compliance engineering (2026) for the cybersecurity control frameworks and product-security regulations that drive architecture, code, and CI gates — not the organizational policy binder. Use when work must satisfy or be audited against NIST CSF 2.0,
name: sota-security-compliance description: >- State-of-the-art security & compliance engineering (2026) for the cybersecurity control frameworks and product-security regulations that drive architecture, code, and CI gates — not the organizational policy binder. Use when work must satisfy or be audited against NIST CSF 2.0, SP 800-53, SP 800-171 / CMMC, the Secure Software Development Framework (SSDF, SP 800-218), FedRAMP, the EU Cyber Resilience Act (CRA), or ISA/IEC 62443 (OT/ICS/embedded). Covers control-framework-as-code crosswalks (control → engineering mechanism → evidence), CUI boundaries, FIPS-validated crypto, SBOM + coordinated vulnerability disclosure + security-update obligations, secure-SDLC gates, and OT zones/conduits & security levels. Complements sota-privacy-compliance (personal data, GDPR, SOC 2, ISO 27001). Trigger keywords: compliance, NIST, CSF, 800-53, 800-171, CMMC, CUI, SSDF, FedRAMP, CRA, Cyber Resilience Act, SBOM, VEX, CVD, IEC 62443, OT security, ICS, security levels, zones and conduits, FIPS 140.
The engineering half of cybersecurity regulation: how to make a control framework a property of the codebase, pipeline, and infrastructure rather than a binder of policy documents. A control that lives in a schema, an IAM policy, a CI gate, or a signed SBOM survives staff turnover and an assessor's sampling; a control that lives in a wiki page does not.
This skill is the **security/regulation counterpart** to `sota-privacy-compliance` (which owns personal-data lifecycle, GDPR, SOC 2, ISO 27001). It owns the cybersecurity *control frameworks* and *product-security regulations* — NIST CSF 2.0, SP 800-53, SP 800-171 / CMMC, SSDF (SP 800-218), FedRAMP, the EU Cyber Resilience Act, and ISA/IEC 62443 for OT.
> **This is engineering guidance, not legal, certification, or assessment advice.** > Framework versions, regulation dates, and conformity routes move — every date > and version below was verified against a primary source (NIST CSRC, EUR-Lex, the > Federal Register, ISA/IEC) as of **July 2026** and each rules file cites it. > Re-verify before you rely on a deadline, and route scoping/attestation decisions > (is this system in CUI scope? which CMMC level? is our product "important" under > the CRA?) to your assessor, sponsor, or counsel. This skill tells you how to > build the machinery those decisions require.
Every framework here mixes **engineering controls** with **organizational controls**. This skill owns only the first; it names the second and routes it out, so you neither skip it nor pretend code can satisfy it.
| In scope (drives code / architecture / gates) | Out of scope (org/governance — note & route out) | |---|---| | System/authorization boundary definition, segmentation, zones & conduits | Security-awareness training programs, phishing simulations | | FIPS-validated crypto selection; encryption at rest/in transit | Personnel screening, background checks, HR onboarding | | Audit-log families, retention, tamper-evidence | Physical & environmental security (badges, data-center access) | | SBOM generation, coordinated vuln disclosure, signed update channels | Risk-committee structure, CISO/DPO roles, board reporting | | Secure-SDLC gates (SAST/DAST/provenance) as CI checks | Written policy authorship, management review cadence | | Control-as-code mappings + machine-generated evidence | Legal interpretation, contract clauses, insurance |
When a framework requirement is organizational, the finding is *"owned by GRC/HR/ legal — here is the engineering hook (an event, an export, an enforcement point) that makes it auditable,"* not silence and not a code change that pretends to cover it.
**Related skills — reference, do not duplicate:**
When designing or implementing a system that must satisfy a framework:
1. **Scope the boundary first.** Before controls, decide what is *in* the assessed system — the authorization boundary (FedRAMP), CUI enclave (800-171/CMMC), or zone (62443). A tight, honest boundary is the cheapest control decision you will make; a flat network drags the whole estate into scope (rules/02, rules/05). 2. **Pick the framework spine, then crosswalk down.** Use NIST CSF 2.0 as the organizing map, then bind each outcome to a concrete mechanism from an existing sota-* skill and record it as code (rules/01). Do not re-implement encryption, RBAC, or logging here — reference where they already live. 3. **Encode controls as gates, not prose.** Encryption-required, FIPS-crypto-only, boundary-egress-deny, SBOM-present, no-known-exploitable-vulns-at-release become policy-as-code and CI checks that fail the build (rules/01 §evidence, rules/03). 4. **Build product-security obligations in from day one** where the CRA/62443 apply: SBOM in CI, a coordinated-vulnerability-disclosure intake, a signed update channel, and a defined support period are architecture, not paperwork — brutal to retrofit (rules/
Make your AI coding assistant build and audit like your most senior engineer. Your assistant is brilliant — it just doesn't know your standards, and it forgets the ones it does know as the task grows long.
Repo: martinholovsky/SOTA-skills
State-of-the-art API design and audit guidance (2026) covering REST/HTTP, GraphQL, gRPC, WebSockets/SSE/realtime, webhooks, versioning/evolution, and API…
State-of-the-art software and system architecture rules (2026) for both building and auditing. Use when designing, building, refactoring, or extending system…
State-of-the-art rules for writing and auditing asynchronous and concurrent code across runtimes (Python asyncio, JS/Node, Go, Rust, JVM). Use when building…
State-of-the-art C and C++ engineering rules (2026 baseline) that Claude applies when writing or auditing C/C++. Covers modern idioms (RAII, value semantics,…
State-of-the-art CLI and developer-tool UX guidance (2026) covering command and flag design, output and interaction (stdout/stderr, --json, TTY detection, exit…
State-of-the-art cloud infrastructure architecture (2026). Applies when designing, building, or auditing cloud environments on AWS, GCP, or Azure —…