skills-as-a-mcp-service

Governs AI models through inventory, purpose, risk tier, approval, monitoring, drift, kill-switch and retirement. Use when introducing, changing or operating models in production or agent workflows.

Applies an Asimov-inspired hierarchy of AI safety laws when material AI or instruction risk is present. Use when harm, unlawful requests, authority conflicts, or safety trade-offs need an explicit law hierarchy. Also known as apply_laws_of_AI.

Use when implementing input validation, output filtering, tool restrictions, policy enforcement, secure coding controls, destructive-tool guardrails, approval gates, or safe agent behaviour.

Use when starting a session or deciding which local skill applies; acts as the meta-skill for discovering and loading the smallest relevant skill.

Use when measuring coding-agent quality, RAG evaluation, regression risk, latency, cost, reliability, safety, or production performance for agent systems.

Designs ontology-driven knowledge graphs for explainable KG-RAG. Use when modelling semantic graphs, ontologies, or graph schemas for retrieval.

Applies SRE practices for production reliability, service ownership, resilience, toil reduction and operability. Use when owning production services, defining SLOs, or improving operational maturity.

Applies WCAG 2.2 AA-aligned accessibility checks to interfaces, forms, dashboards and components. Use when building or reviewing any user-facing UI, content, or agent supervision screen.

Use when designing or changing build, test or deployment automation that needs explicit safety, permissions and verification.

Use when reviewing code before merge to assess correctness, tests, maintainability, security and user impact.

Use when a change needs durable documentation, an architecture decision record or maintainership guidance.

Designs event-driven systems with asynchronous flows, brokers, producers, consumers and contracts. Use when architecting asynchronous, decoupled, or event-first systems, including order status updates and domain event flows.

Use when a coding agent must request approval, expert judgement, clarification, review, or intervention before proceeding with high-risk or ambiguous work.

Handles incidents, recovery, postmortems and corrective actions with blameless, evidence-led practice. Use when responding to outages, writing postmortems, or defining incident runbooks.

Designs actionable logs, metrics, traces, dashboards and alerts for reliable operations. Use when instrumenting services, defining SLIs, or improving on-call signal quality.

Use when a request is underspecified and the user's actual goal, constraints or success criteria must be elicited before implementation.

Applies NIST SSDF (SP 800-218) secure development and supply-chain controls including SBOM, pinning, secrets, SAST and MCP/A2A integrity. Use when hardening builds, dependencies, CI/CD, releases or agent tool supply chains.

Use when a feature or significant change needs a decision-complete specification before implementation begins.

Applies TDD: failing test first, smallest passing change, refactor under test protection. Use when adding behaviour, fixing defects, or refactoring with executable expectations.

Identifies assets, trust boundaries, threats and mitigations for systems and agents. Use when designing security-sensitive features, reviewing attack surface, or planning abuse-case tests.

Applies UX principles for clear, task-centred journeys, navigation and forms. Use when designing new screens, workflows, information architecture, or user journeys.

Use when implementing retrieval-augmented generation, RAG features, RAG pipelines, document-grounded coding agents, semantic search, citation-backed answers, repository/document knowledge lookup, evaluation gates, or release evaluation for RAG.

Evaluate and govern KRAG and GraphRAG systems only. Use when defining KRAG-scoped acceptance thresholds, test design, observability, drift, grounding evidence, and KRAG promotion blockers.