ch4570

@ch4570 source repo

76 skills
0 followers
1 day ago last updated

▌
Bandit · ch4570 bundle
Coordinate product planning with BANDIT for general or combined research, MVP scope, marketing strategy, product design handoff, business model (BM) design, specification creation or rewriting, and review. Use a named BANDIT specialist directly when selected; excludes standalone UI styling, code implementation, and personal task scheduling.

0
installs
▌
Bandit Scope · ch4570 bundle
Define MVP scope and feature priorities under time, team, and product constraints. Use to choose what to build now, defer, or handle manually and compare product alternatives; excludes personal task scheduling and code implementation.

0
installs
▌
Bandit Review · ch4570 bundle
Review existing product plans and PRDs for consequential gaps, conflicting rules, unsupported claims, and verification limits, including consistency across marketing promises, design flows, and business model assumptions. Use for scoped planning critique and proposed corrections; leaves the reviewed artifacts unchanged and excludes standalone UI or code-quality audits.

0
installs
▌
Bandit Specify · ch4570 bundle
Create, rewrite, or refine PRDs and product specifications from an idea, changed direction, or new evidence, including requested marketing, UX/design, and business model handoffs. Read an existing spec first and replace outdated rules with a coherent current plan, including flows, permissions, states, and acceptance scenarios; excludes code implementation and standalone visual styling.

0
installs
▌
Bandit Research · ch4570 bundle
Research product opportunities, customer problems, competitors, business assumptions, and demand experiments. Use when the user asks what evidence supports a product decision or what to test next; excludes unrelated general research and implementation.

0
installs
▌
UX Writing · ch4570 bundle
Write or refine truthful product microcopy for action labels, field help, errors, empty states, loading, and confirmations, including natural Korean. Use when interface wording is vague, stiff, misleading, too long, or hard to recover from; preserve behavior, localization contracts, and product terminology.

0
installs
▌
UI Slop Audit · ch4570 bundle
Diagnose generic, cluttered, or AI-looking product UI through read-only evidence and prioritized findings. Use when asked to audit or critique a screen before editing; identify task friction, hierarchy problems, and misleading states without redesigning or changing code.

0
installs
▌
UX Flow Refine · ch4570 bundle
Improve product interaction behavior, immediate feedback, state transitions, error recovery, and purposeful motion. Use when saves feel unreliable, flows are confusing, input or animation races occur, loading obscures progress, or keyboard and touch access fail; preserve the existing data and navigation architecture.

0
installs
▌
AI Slop Remover · ch4570 bundle
Remove generic AI-generated UI and make product experiences clear, calm, and easy to use. Use when asked to remove AI slop, make UI/UX smoother or more user-friendly, or improve a screen across visual design, interaction, and copy. Coordinate only the relevant specialist skills; exclude backend cleanup, prose detection, and standalone illustration.

0
installs
▌
UI Craft Bundle · ch4570 bundle
Preserve the original UI Craft Bundle workflow and provide shared web/native design, interaction, motion, and verification references. Use when the user explicitly requests ui-craft-bundle or an existing workflow depends on its references. For a new broad AI-slop-removal request, prefer ai-slop-remover when installed; exclude backend-only work and standalone graphic design.

0
installs
▌
UI Quality Gate · ch4570 bundle
Verify a completed UI change through a read-only review of scoped visual and behavioral evidence. Use for final UI/UX QA, completion checks, or release-readiness assessment after implementation; identify concrete task, accessibility, layout, state, and evidence gaps without changing product code or relying on beauty scores.

0
installs
▌
UI Visual Refine · ch4570 bundle
Refine a product screen's color composition, layout, hierarchy, grouping, spacing, typography, density, and responsive structure while preserving its brand and behavior. Use for weak palettes or placement, visual cleanup, inconsistent layouts, cramped content, or narrow styling fixes; leave workflow logic and copy rewrites to their specific skills.

0
installs
▌
Opensearch Code Authoring · ch4570 bundle
Write or modify OpenSearch mappings, Query DSL, vector or hybrid search settings, and reindex plans with relevance, latency, and operational safety. Use when a feature, bug fix, or refactor in a repository detected as OpenSearch touches search behavior, mapping, configuration, or queries, even when the user does not name OpenSearch; do not use for review-only requests.

0
installs
▌
Opensearch Review Workflow · ch4570 bundle
Run the complete OpenSearch review suite for queries, mappings, vector/search tuning, and operational safety, then synthesize one evidence-based verdict. Use for OpenSearch or Elasticsearch-compatible index, query, migration, relevance, capacity, or performance changes.

0
installs
▌
Java Spring Review Workflow · ch4570 bundle
Review Java and Spring changes by running the exact Java/Spring correctness reviewer, object-oriented design review, and design-pattern fit review concurrently against one frozen scope, then reconcile them into one evidence-based verdict. Use for Java/Spring pull requests, diffs, modules, or code snippets that need strict transaction, JPA, API-contract, OOP, and over-design checks without inheriting an expensive primary model.

0
installs
▌
Architecture Review Workflow · ch4570 bundle
Review MSA, modular-monolith, and multi-module boundaries by combining system architecture, application architecture, DDD, and code-topology evidence. Use for service decomposition, module extraction, dependency-direction, data ownership, integration contracts, or architecture changes that need boundary-by-boundary findings.

0
installs
▌
Backend Code Review Workflow · ch4570 bundle
Orchestrate a consolidated backend review across Kotlin/Spring correctness, refactoring, design patterns, object-oriented design, and security. Use for pull requests, diffs, or backend modules that need one severity-gated report without losing specialist dissent.

0
installs
▌
Kotlin Spring Review Workflow · ch4570 bundle
Run Kotlin/Spring, object-oriented design, design-pattern, and behavior-preserving refactoring reviews against one frozen code scope, then reconcile them into one evidence-based verdict. Use for Kotlin changes, including Kotlin/Spring pull requests, diffs, modules, or code snippets that need a multi-angle design and maintainability review without the broader security scope of a full backend audit.

0
installs
▌
Test Quality Refactoring Workflow · ch4570 bundle
Freeze a repository test scope, audit test quality, make only authorized Kotlin/JVM test-side improvements, prove fault detection, and reconcile fresh execution evidence without overwriting user changes. Use for a bounded test-quality audit or a safe test refactor; do not use for production fixes, broad cleanup, E2E, or adding test tooling.

0
installs
▌
Enum · ch4570 bundle
Scaffold a protocol enum in the shared SSOT module with the unknown-value fallback convention. Use when adding a domain/protocol enum shared across modules.

0
installs
▌
Learn · ch4570 bundle
Extract reusable, non-obvious insights from the current conversation/work and persist them as team-shared knowledge under .claude/knowledge/ (rule changes are proposed against AGENTS.md or the relevant skill instead). Applies changes only after user approval. Use when a durable lesson, gotcha, domain flow, or decision rationale emerged that future sessions should not have to rediscover.

0
installs
▌
Retro · ch4570 bundle
Work retrospective for this repository. Reviews a finished task/session, structures it as Keep/Problem/Try, and derives concrete harness-improvement proposals (rules, skills, agents, knowledge). Persists the retro log under .claude/retro/ and applies approved improvements only after user confirmation. Use after finishing a task or session, or when the user asks to reflect on recent work.

0
installs
▌
Entity · ch4570 bundle
Scaffold a new JPA entity following project conventions (schema-qualified @Table, composite @EmbeddedId keys, Persistable for JDBC upsert, read-only mirror entities). Use when adding a persisted entity.

0
installs
▌
Flyway · ch4570 bundle
Scaffold a Flyway migration SQL file with a build-enforced filename and PostgreSQL conventions. Use when a persistence change (entity/column/index) needs a migration — authored in the SAME change as the entity.

0
installs
▌
Mapper · ch4570 bundle
Scaffold a manual `object` mapper between a domain model and JPA entity. This project maps by hand, not MapStruct, because entities are immutable. Use when adding model-to-entity or entity-to-model conversion.

0
installs
▌
PDF QA · ch4570 bundle
Render every page of an existing PDF to PNG and gate delivery on recorded visual inspection. Use after generating or changing a PDF, before claiming a PDF report, document, diagram, or slide deck is complete, or when diagnosing clipping, overlap, broken glyphs, pagination, page-balance, or table-spacing defects. This skill validates PDFs; it does not author or repair their source.

0
installs
▌
Request · ch4570 bundle
Scaffold a REST request DTO co-located with its controller, as a validated Kotlin data class. Use when adding a request body for an API endpoint.

0
installs
▌
Service · ch4570 bundle
Scaffold a domain service in a domain module — the external entry point for an entity's reads/writes. Collaborators (batch writers, Kafka consumers, API services) call the service, never repositories directly. Use when adding domain logic / a read+write entry point for a domain.

0
installs
▌
Git Flow · ch4570 bundle
Prepare a safe work branch or commit, push, and open a GitLab merge request when the user asks to start, land, or publish implementation work.

0
installs
▌
Response · ch4570 bundle
Scaffold a REST response DTO in the API module with a `from(model)` companion factory; paginated lists use the shared cursor page envelope. Use when adding a response body for an API endpoint.

0
installs
▌
Agent Eval · ch4570 bundle
Evaluate a new or changed agent or skill before merge with NVIDIA SkillEvaluator evidence and adversarial review.

0
installs
▌
E2e Runner · ch4570 bundle
Run repository E2E or integration tests in an isolated Testcontainers environment; use for E2E, integration, QA, or full test requests.

0
installs
▌
Repository · ch4570 bundle
Scaffold data access for a domain entity — a JpaRepository interface named *EntityRepository, plus optional *JdbcRepository (bulk ON CONFLICT upsert) and *AggregateRepository (QueryDSL). Use when adding persistence access.

0
installs
▌
Start Task · ch4570 bundle
Finish implementation or review directly by default, escalating for uncertainty or risk. Load this entry together with independent reads of known task files.

0
installs
▌
Nl SQL Query · ch4570 bundle
Use when a user asks a data question in natural language and wants the SQL built, executed (read-only), and results shown — for PostgreSQL/MySQL/MS-SQL via the nl-sql MCP. Explore the schema with tools, then author and run a dialect-appropriate SELECT and present the results.

0
installs
▌
Vulpora Init · ch4570 bundle
Initialize or refresh repository-level Vulpora routing guidance in AGENTS.md from detected technology-stack evidence. Use immediately after installing Vulpora skills, when opening a repository for the first time, or after build dependencies and persistence/search infrastructure change; do not use for ordinary feature implementation.

0
installs
▌
Skill Updater · ch4570 bundle
Create or update a portable Vulpora skill when adding, editing, or validating a SKILL.md package.

0
installs
▌
Diagram Styler · ch4570 bundle
Render an existing Mermaid `.mmd` semantic source into a themed, responsive, accessible SVG while preserving the source and recording renderer provenance in a manifest. Use for light or dark Mermaid styling, document-width SVG output, presentation output, and pre-rendering diagrams for HTML or PDF documents. Do not use to change diagram meaning or to download renderers at runtime.

0
installs
▌
Playwright E2e · ch4570 bundle
Execute QA-approved browser E2E in a run-owned Node Testcontainers environment. Prove the app/baseURL and stateful endpoints are ephemeral, seed synthetic case data, execute once, verify cleanup, teardown with zero orphans, and report evidence. Use after QA hands off UI scenarios or when asked to run Playwright E2E.

0
installs
▌
Test Authoring · ch4570 bundle
Write or review trustworthy Kotlin unit and narrow integration tests for agent-authored changes. Prefer real contracts; exclude E2E and live-service tests.

0
installs
▌
Harness Propose · ch4570 bundle
Research harness-engineering trends against the project's actual capabilities and write a self-contained HTML improvement proposal. Use for harness improvement proposals or periodic harness reviews, including projects without an existing report. Apply changes only within user-authorized scope.

0
installs
▌
Knowledge Audit · ch4570 bundle
Audit the autonomous knowledge lane (.claude/knowledge/auto/) — verify each machine-written candidate against current code, then promote true/reusable ones to the curated knowledge base or discard stale/duplicate/wrong ones. This is the curation half of the autonomous learning loop — it replaces the human approval gate that the Stop auto-learn hook bypasses. Use periodically (or when the auto lane has accumulated candidates) to keep the KB from rotting.

0
installs
▌
Mermaid Diagrams · ch4570 bundle
Create or revise editable Mermaid source for version-controlled flowcharts, sequence diagrams, state diagrams, ER diagrams, class diagrams, and simple architecture diagrams. Use when a structural relationship, process, interaction, lifecycle, data model, or small component topology is clearer as a diagram than prose or a table. Do not use for whiteboard aesthetics, pixel-precise branded layouts, charts, or cases where text is clearer.

0
installs
▌
Test Refactoring · ch4570 bundle
Refactor selected Kotlin/JVM tests without weakening their fault-detection evidence. Test and fixture files only.

0
installs
▌
Document Designer · ch4570 bundle
Turn supplied content into a polished, accessible, editable Markdown document with clear structure, typography roles, information hierarchy, tables, callouts, captions, and page-aware layout intent. Use for reports, PRDs, technical documents, memos, and other Markdown-first publications before HTML or PDF rendering. Do not use to render HTML/PDF or to invent missing content.

0
installs
▌
E2e Test Workflow · ch4570 bundle
Orchestrate E2E catalog authoring, API or integration execution, Playwright browser execution, and optional HTML reporting under one run-owned Testcontainers lifecycle. Use when the agent must run a complete or multi-lane E2E verification, refresh and execute the scenario catalog, combine backend and browser evidence, or produce one false-green-safe E2E handoff. Do not use for authoring-only, runner-only, browser-only, or render-only requests that one component skill can satisfy directly.

0
installs
▌
Explore Dead Code · ch4570 bundle
Quickly find unreferenced dead-code candidates with file/line evidence and confidence. Use for unused functions, classes, exports, files, or 참조 없는 죽은 코드 탐색. Read-only discovery; excludes code deletion and general refactoring.

0
installs
▌
Korean Dev Writer · ch4570 bundle
한국어 기술 문서, PR, 리뷰, 코드 주석을 사실과 요구사항을 보존하며 자연스럽고 간결하게 작성·교정한다. 일반 문예 글쓰기나 새 사실 작성에는 사용하지 않는다.

0
installs
▌
Oop Design Review · ch4570 bundle
Review object responsibilities, cohesion, encapsulation, and collaboration using SOLID and GRASP. Use for object-oriented design review, not implementation scaffolding.

0
installs
▌
Product UI Design · ch4570 bundle
Design, implement, or review web, mobile, and desktop product interfaces from real product context and the repository's existing design language. Use for pages, flows, components, responsive behavior, interaction states, accessibility, and visual refinement. Reject generic template aesthetics and unverified visual-completion claims. Do not use for document, slide, or technical-diagram styling.

0
installs
▌
Vulpora Installer · ch4570 bundle
Install, update, verify, or remove Vulpora agents, skills, and MCP packs for Codex or Claude Code. Use when the user asks to set up Vulpora, install a catalog asset, remove an installed asset, or connect the Notion/OpenAI Docs MCP pack. Execute requested changes through a checkout, installed CLI, or the public repository npx launcher; reserve command handoff for unavailable tools or explicitly interactive requests. Do not use this skill for Notion research itself.

0
installs
▌
Markdown Publisher · ch4570 bundle
Markdown 문서를 편집 가능한 원본과 함께 오프라인 단일 HTML로 발행하고, 요청 시 로컬 Chromium 계열 브라우저로 PDF를 생성한다. 기술 문서, PRD, 메모, 보고서에 executive/technical/minimal 테마를 적용하거나 Markdown을 인쇄 가능한 HTML/PDF로 변환할 때 사용한다. PDF 완료 판정과 페이지별 시각 검수는 pdf-qa로 라우팅한다.

0
installs
▌
Schema Doc Extract · ch4570 bundle
Generate or refresh a Mermaid ERD and per-schema table specifications from migration SQL and ORM code. Static extraction only; no live database.

0
installs
▌
Codex Agent Runtime · ch4570 bundle
Record the fail-closed compatibility status of legacy Vulpora catalog-agent execution. This installable negative-contract skill must not run agents, discovery probes, or delegated tasks.

0
installs
▌
E2e Report Renderer · ch4570 bundle
Render settled e2e-runner JSON as offline HTML in single-run, trend, or comparison mode when asked to visualize or share E2E results. Preserve input data.

0
installs
▌
E2e Scenario Author · ch4570 bundle
Generate or refresh a source-validated E2E scenario catalog for Spring JVM or NestJS after endpoints/listeners change. Authoring only; does not execute tests.

0
installs
▌
Postgres Risk Check · ch4570 bundle
Check the risk of a PostgreSQL change (migration/DDL/bulk DML). Inspect DDL lock levels, zero-downtime change patterns (CONCURRENTLY, NOT VALID→VALIDATE), chunking of large backfills, transactions/concurrency/deadlocks, lock/statement timeout, bloat/VACUUM, and rollback/downtime. Use PROACTIVELY when authoring/reviewing a migration, before a production deploy, or for destructive changes such as DROP/type change.

0
installs
▌
Refactoring Catalog · ch4570 bundle
Select behavior-preserving refactorings for demonstrated code smells and give an ordered verification plan. Use when deciding how to improve existing code safely.

0
installs
▌
Test Quality Review · ch4570 bundle
Audit whether existing tests detect plausible defects and produce a validated, evidence-backed quality report without changing repository files. Use when the agent must review test oracle strength, fault detectability, boundaries, isolation, destructive cleanup ownership, refactoring resistance, or fresh execution evidence before deciding whether tests need improvement.

0
installs
▌
Code Diagram Extract · ch4570 bundle
Trace a function or feature into a Mermaid flow, sequence, or state diagram with real source-file and line anchors. Use to explain code behavior.

0
installs
▌
Design Pattern Apply · ch4570 bundle
Assess whether a design pattern fits a real axis of change and prescribe its application. Use for pattern selection or composition/inheritance decisions; reject over-design.

0
installs
▌
Kotlin Spring Review · ch4570 bundle
Reference knowledge base for reviewing Kotlin + Spring MSA backend code. Provides DDD three-layered and hexagonal architecture rules, Clean Code principles, Kotlin idioms, functional JVM style, Spring best practices, a severity rubric, and an integrated checklist. Use when reviewing Kotlin/Spring code, checking code quality or architecture rules, or when you need Kotlin-Spring review criteria.

0
installs
▌
Mssql Code Authoring · ch4570 bundle
Write or modify Microsoft SQL Server and legacy MS-SQL queries, stored procedures, schemas, indexes, and migrations with version and compatibility-level gating, parameter safety, transaction correctness, and operational verification. Use when repository evidence shows SQL Server, T-SQL, Microsoft.Data.SqlClient, System.Data.SqlClient, mssql-jdbc, mssql, or tedious and the task changes a query, persistence path, database object, or migration. Do not use for review-only work or for PostgreSQL, MySQL, or generic ANSI SQL without SQL Server evidence.

0
installs
▌
Product Requirements · ch4570 bundle
제품 아이디어, 기능 제안, 사용자 문제를 구현 가능한 제품 요구사항 문서(PRD)로 구체화한다. 제품 범위와 사용자 흐름을 정리하고, 관찰 가능한 기능 요구사항과 Given/When/Then 인수 기준을 연결하거나, 개발 handoff를 준비해야 할 때 사용한다. 아이디어 자체를 구현·배포하거나 외부 시스템을 변경하는 작업에는 사용하지 않는다.

0
installs
▌
Code Authoring Router · ch4570 bundle
Route generic feature, bug-fix, refactor, and implementation requests to installed code-authoring skills after detecting repository stack evidence. Use even when the user only names the desired behavior and does not mention Kotlin, Spring, PostgreSQL, Microsoft SQL Server, or OpenSearch; do not use for review-only requests.

0
installs
▌
Kotlin Code Authoring · ch4570 bundle
Write or modify Kotlin application code while preserving project conventions and readable Kotlin idioms. Use when implementing a feature, bug fix, or refactor in actual Kotlin application source, including Kotlin/Spring modules, even when the user does not name the stack; do not use for Java changes, Gradle build logic, or review-only requests.

0
installs
▌
Notion Domain Context · ch4570 bundle
Retrieve cited, read-only evidence from an explicitly connected private Notion workspace for internal domain or policy questions. Exclude sensitive HR/medical/payroll data, writes, repository-local questions, and MCP setup.

0
installs
▌
Postgres Query Review · ch4570 bundle
Review PostgreSQL queries (SQL) from performance and correctness angles. Inspect execution-plan interpretation, index usability, join method/order, sort/aggregate cost, pagination, parameter binding, and anti-patterns. Use when writing/modifying SQL or diagnosing slow queries. Applies the principles of "Friendly SQL Tuning", verified against the official PostgreSQL documentation.

0
installs
▌
Postgres Schema Design · ch4570 bundle
Design and review PostgreSQL tables/schemas. Covers data modeling (entities·relationships·normalization), identifier strategy (natural/surrogate keys), constraints·integrity (PK/FK/UNIQUE/CHECK), type selection, history modeling, and naming conventions. Use when writing new table/migration DDL or reviewing an existing schema. Applies the principles of "Core Data Modeling," validated against the official PostgreSQL documentation.

0
installs
▌
Security Scan Workflow · ch4570 bundle
Analyze security vulnerabilities and shared-resource risks across a diff, module, or repository. Use for security scans, hardening, or requests to remove unsafe code: authentication, tenant isolation, secrets, browser and interpreter injection, SSRF, files, configuration, dependencies, resource exhaustion, and destructive DB/cache/search/queue/storage operations. Audit read-only when asked to review; when asked to fix, patch verified risks and run safely isolated regression checks with independent re-review. Never operate shared services to demonstrate a flaw.

0
installs
▌
Visual Artifact Router · ch4570 bundle
Route requests for polished Markdown, HTML, PDF, Mermaid, SVG, PNG, and technical diagrams to the smallest suitable visual-artifact workflow. Use when the agent must choose a document or diagram format, preserve editable sources, attach render-based QA, or decide whether a diagram adds value. Do not use for ordinary prose where visual publishing is not requested or useful.

0
installs
▌
Opensearch Optimization · ch4570 bundle
Tune OpenSearch 3.5 vector/search optimization parameters and diagnose operational risk. Covers HNSW/IVF parameters (m/ef_construction/ef_search/nlist/nprobes), quantization (SQ/PQ/BQ)·on_disk, memory estimation· off-heap·circuit breaker, indexing tuning (refresh/replica/force merge)·shard sizing·ISM·caching. Use for performance/memory/recall tuning and migration/operational-safety checks. Grounded in the official-docs KB.

0
installs
▌
Opensearch Query Review · ch4570 bundle
Review OpenSearch 3.5 search queries. Inspect dynamic bool queries (filter vs query context, builder safety, paging), vector k-NN queries (k/ef_search, efficient filtering, under-fetch, radial, nested), hybrid (normalization/RRF)·neural/sparse/semantic, and search pipelines. Use when authoring/modifying queries or diagnosing search-result/latency problems. Ground claims in the OpenSearch official-docs KB.

0
installs
▌
Postgres Code Authoring · ch4570 bundle
Write or modify PostgreSQL queries, schemas, and migrations with parameter safety, data integrity, and operational verification. Use when a feature, bug fix, or refactor in a repository detected as PostgreSQL requires SQL, DDL, indexes, or migrations, even when the user does not name the database; do not use for review-only requests.

0
installs
▌
Opensearch Schema Review · ch4570 bundle
Review OpenSearch 3.5 index mappings/schemas. Check the dynamic policy, text/keyword, multi-field, index/doc_values, _source/derived source, nested cost, knn_vector fields (dimension/ space_type/engine/method/mode), and the Korean nori analyzer. Use when designing/modifying a mapping or reviewing an index schema. Grounds findings in a KB based on the official OpenSearch docs.

0
installs
▌
Postgres Review Workflow · ch4570 bundle
Run all PostgreSQL query, migration-risk, and schema-design skills plus operational DBA and data modeling reviewers, then reconcile their advice. Use for SQL, DDL, migrations, data models, performance incidents, or database changes that need one deployability-aware verdict.

0
installs

76 published skills