kalilurrahman/kr-claudiator-skills - 第 13 页

data-lineage

Design and implement data lineage tracking across a data pipeline. Covers lineage capture strategies, metadata graphs, column-level lineage, impact analysis, and integration with dbt, Airflow, and OpenLineage.

原文语言：英语

data-migration

Design and execute safe database migrations with zero downtime, rollback plans, and data validation. Outputs migration scripts, expand-contract patterns, backfill strategies, and cutover runbooks.

原文语言：英语

data-partitioning

Design database and data lake partitioning strategies for scale, performance, and cost optimization. Outputs partitioning schemes, partition pruning patterns, maintenance procedures, and query optimization.

原文语言：英语

data-retention

Design data retention policies covering deletion schedules, archival tiers, compliance requirements, and right-to-erasure workflows. Outputs retention schedules, automated deletion jobs, anonymization strategies, and audit trails.

原文语言：英语

etl-pipeline

Design and implement production-ready ETL/ELT pipelines for data warehousing and analytics. Covers extraction strategies, transformation logic, incremental loading, data quality, orchestration, schema evolution, and monitoring.

原文语言：英语

feature-store

Design and implement a feature store for ML — centralized feature computation, storage, serving, and reuse across models. Outputs online/offline store architecture, feature pipelines, point-in-time joins, and serving API.

原文语言：英语

fine-tuning

Fine-tune a pre-trained language model for a specific task. Covers dataset preparation, LoRA/QLoRA techniques, training configuration, evaluation, merging, and deployment of fine-tuned models.

原文语言：英语

llm-ops

Operationalize LLM-powered applications in production — covering prompt versioning, evaluation pipelines, cost tracking, latency monitoring, hallucination detection, and safe rollout of model upgrades.

原文语言：英语

synthetic-data

Generate synthetic datasets for ML training, testing, and privacy-preserving data sharing. Outputs statistical generators, GAN-based synthesis, differential privacy techniques, and validation pipelines.

原文语言：英语

vector-database

Design and implement a vector database for semantic search, RAG, and similarity lookups. Covers embedding strategy, index selection, query patterns, metadata filtering, and performance tuning.

原文语言：英语

incident-response

Execute a structured security incident response from detection through post-incident review. Covers the PICERL lifecycle, severity classification, containment playbooks, evidence preservation, communication templates, and blameless post-mortems.

原文语言：英语

penetration-test-plan

Plan a penetration test engagement for a web application, API, or infrastructure target. Produces scope definition, test methodology, rules of engagement, finding classification, and remediation SLAs.

原文语言：英语

security-headers

Configure HTTP security headers to protect against XSS, clickjacking, MIME sniffing, and other browser-based attacks. Outputs CSP policies, HSTS, X-Frame-Options, and scanning/testing configuration.

原文语言：英语

supply-chain-security

Secure the software supply chain against dependency attacks, compromised pipelines, and malicious packages. Covers SBOM generation, dependency pinning, artifact signing, SLSA framework, secret scanning, and private registry configuration.

原文语言：英语

zero-trust

Design and implement zero-trust network architecture — never trust, always verify. Outputs identity-aware proxy config, mTLS policies, device trust enforcement, micro-segmentation rules, and continuous verification pipelines.

原文语言：英语

api-versioning

Design an API versioning strategy for a REST or GraphQL API. Covers version negotiation, deprecation lifecycle, migration guides, and backward compatibility rules.

原文语言：英语

cdc-patterns

Design a Change Data Capture (CDC) pipeline to stream database changes to downstream consumers. Outputs connector config, schema design, consumer patterns, ordering guarantees, and failure handling.

原文语言：英语

cqrs-pattern

Implement Command Query Responsibility Segregation separating read and write models. Outputs command handlers, query handlers, read model synchronization, and eventual consistency patterns.

原文语言：英语

database-sharding

Design a horizontal database sharding strategy to scale beyond the limits of a single database node. Covers shard key selection, sharding strategies, cross-shard queries, rebalancing, and operational challenges.

原文语言：英语

load-balancing

Design load balancing architecture — algorithms, health checking, session affinity, layer 4 vs layer 7, geographic distribution, and failover strategies for high-availability services.

原文语言：英语

multi-tenancy

Design multi-tenant SaaS architecture with tenant isolation, data partitioning, resource quotas, and tenant-aware observability. Outputs isolation patterns, row-level security, schema provisioning, and quota enforcement.

原文语言：英语

saga-pattern

Design distributed transactions using the Saga pattern to maintain data consistency across microservices without two-phase commit. Covers choreography vs orchestration, compensating transactions, failure handling, and implementation examples.

原文语言：英语

mutation-testing

Use mutation testing to measure and improve the quality of a test suite. Covers mutation operators, interpreting mutation scores, fixing weak tests, and integrating mutation testing into CI with Stryker, PIT, or mutmut.

原文语言：英语

test-data-management

Design and implement a test data strategy that provides consistent, realistic, and isolated data for all test levels. Covers factories, fixtures, seeding, anonymisation, synthetic data, and database-per-test patterns.

原文语言：英语

test-driven-development

Apply Test-Driven Development (TDD) to design and build production code. Outputs test suites, refactored implementations, and coverage strategies using the Red-Green-Refactor cycle.

原文语言：英语

event-driven-api

Design an event-driven API strategy choosing between webhooks, Server-Sent Events, WebSockets, and polling. Covers delivery guarantees, retry logic, security, fan-out patterns, and tool selection.

原文语言：英语

openapi-design

Design and document REST APIs using the OpenAPI 3.1 specification — schemas, security definitions, response codes, examples, and SDK generation. Outputs complete OpenAPI specs, documentation, and validation tooling.

原文语言：英语

ab-test-design

Design a statistically valid A/B test for a product feature or change. Produces a test plan with hypothesis, metrics, sample size calculation, risk assessment, and analysis plan.

原文语言：英语

competitor-scan

Conduct a structured competitive analysis of a product or feature set. Produces a comparison matrix, positioning gaps, competitive strengths and weaknesses, and strategic recommendations.

原文语言：英语

decision-doc

Write a structured decision document (RFC, ADR, or decision memo) that captures context, options considered, tradeoffs, and rationale. Produces durable, searchable records of why key decisions were made.

原文语言：英语

feedback-analyzer

Systematically analyze product feedback from multiple sources — support tickets, NPS responses, app reviews, user interviews, sales calls — into prioritized themes with evidence, signal strength ratings, and product action recommendations.

原文语言：英语

go-to-market

Plan and execute a product go-to-market strategy covering ICP definition, channel selection, messaging, launch sequencing, and success metrics. Covers PLG, sales-led, and hybrid motions.

原文语言：英语

launch-checklist

Generate a comprehensive product launch checklist covering engineering, design, marketing, legal, support, and analytics readiness. Outputs a trackable launch plan with owners, dependencies, and go/no-go criteria.

原文语言：英语

market-sizing

Estimate the size of a market using top-down and bottom-up methods. Produces a TAM/SAM/SOM analysis with assumptions documented, sensitivity analysis, and business implications for go-to-market strategy.

原文语言：英语

meeting-prep

Prepare for a product management meeting, including stakeholder alignment meetings, roadmap reviews, planning sessions, or executive updates. Produces an agenda, pre-read materials, and a facilitation guide.

原文语言：英语

metrics-analyzer

Analyze a product or business metric, diagnose root causes, and produce a structured investigation report with hypotheses, supporting data, and recommended actions.

原文语言：英语

okr-writer

Write clear, measurable OKRs for product teams — objectives that inspire, key results that measure outcomes, and initiative mapping that connects work to results. Outputs draft OKRs with grading rubrics, check-in templates, and failure pattern analysis.

原文语言：英语

persona

Create evidence-based user personas grounded in research — not fictional archetypes. Outputs persona profiles with jobs-to-be-done, behavioral patterns, pain points, workflow context, and product decision implications.

原文语言：英语

prd

Write a complete Product Requirements Document from a feature idea or problem statement. Produces a structured PRD with goals, user stories, acceptance criteria, edge cases, and success metrics.

原文语言：英语

pricing-strategy

Design a product pricing strategy covering model selection, tier structure, value metric identification, packaging, and pricing research. Covers freemium, usage-based, per-seat, and enterprise packaging with worked examples.

原文语言：英语