Srltas
UserMy Claude Skills
Categories
Indexed Skills (22)
commit
Commit staged changes with a clean, single-line Conventional Commit subject in plain developer English. Use when the user wants to commit changes, write a commit message, or wrap up work with a conventional-commit title. Generates a `type(scope): summary` subject from the staged diff and commits it. Triggers on phrases like 'commit this', 'make a commit', 'write a commit message', 'git commit', 'commit my changes'.
cubrid-manual
Look up CUBRID engine behavior in the official online manual: SQL syntax, built-in functions, data types, identifiers and reserved keywords, configuration parameters, and JDBC/driver APIs. Fetches the version-pinned manual at www.cubrid.org/manual. Use when you need to verify CUBRID SQL semantics, a function signature, a data type's range, whether a word is reserved, or a config parameter before writing dialect or JDBC code. For CUBRID Migration Toolkit (CMT) questions, use the cmt-manual skill instead. Triggers on phrases like 'how does X work in CUBRID', 'CUBRID syntax for', 'is X a reserved word in CUBRID', 'what functions does CUBRID have', 'check the CUBRID manual'.
hhh-dialect-verify-report
One-shot CUBRID dialect verification + report. Runs the hibernate-core test suite against one or ALL CUBRID versions, diffs pass/fail vs a baseline (a prior run's JSON, results dir, or .tgz), and ALWAYS produces a Korean Word (.docx) 비교 분석 report. Use when you want the full 'verify → measure → report' pipeline for a CUBRIDDialect change in one go, or to test across all CUBRID versions. Composes the hhh-dialect-verify and report skills. Triggers on phrases like 'dialect 검증하고 리포트까지', 'CUBRID 전체 버전 테스트 + 보고서', 'verify the dialect and write the report', 'run all CUBRID versions and report'.
hhh-dialect-verify
Run the Hibernate ORM test suite against CUBRID and measure a dialect change's impact: total failures, +N recovered / −N regressed vs a baseline, and failures grouped by signature/family. Use when verifying a CUBRIDDialect change, an auto_quote or capability-flag toggle, or any Hibernate-vs-CUBRID compatibility measurement. Pairs with the report skill (verify → numbers → 비교 분석 보고서). Triggers on phrases like 'dialect 검증', 'CUBRID 테스트 돌려', 'hibernate 스위트 실행', 'pass/fail 델타 측정', 'run the hibernate suite against CUBRID', 'measure dialect change'.
jdbc-ctp-verify-report
One-shot CUBRID JDBC driver verification + report. Builds the CUBRID JDBC driver, runs the CTP JDBC test suite against CUBRID, diffs pass/fail vs a baseline (a prior run's JSON, CTP result dir, or test-jdbc.xml), classifies broken cases (VerifyError, conversion, broker/env, …), and ALWAYS produces a Korean Word (.docx) 비교 분석 report. Use to verify a JDBC driver change or a Java-version/build-target change (e.g. v50 1.6 vs v52 1.8) end to end. Composes parse_ctp.py + the report skill. Triggers on phrases like 'JDBC 드라이버 검증', 'CTP 돌려', 'ctp jdbc 실행하고 리포트', 'verify the JDBC driver with CTP', 'build driver and run CTP'.
jira-draft
Draft a CUBRID JIRA issue (bug or task) with a concise English title and a Korean body under English section headers (Description as easy-to-read prose; the other sections in bullet style, 개조식). Use when you need to write up a CUBRID JIRA issue from the current work: a bug report (Description / Test Build / Repro / Expected / Actual / Additional Info) or a task/improvement (Description / Specification Changes / Implementation / Acceptance Criteria / Definition of Done). Produces a copy-paste draft only; it does not post to JIRA. It is the reverse of jira-fetch (which downloads an issue). Triggers on phrases like 'CUBRID JIRA 이슈 작성', 'jira 버그 리포트 초안', '이 작업 jira 이슈로 정리', 'draft a CUBRID jira issue', 'write a jira bug report'.
jira-fetch
Download a JIRA issue to a local Markdown file to use as working context, via the Srltas/jira-to-md-downloader tool. Give it one or more issue keys or issue URLs (e.g. CBRD-1234, TOOLS-4888, APIS-1079, or a pasted http://jira.cubrid.org/browse/CBRD-1234) and it fetches each issue's summary + description as <KEY>.md, then loads it into the session so you can work against the ticket. Built for Jira Server/Data Center (REST API v2 + HTTP Basic auth); the target is CUBRID's Jira Server at jira.cubrid.org, and it reaches the publicly readable projects (CBRD, TOOLS, APIS, ...) only: internal-only projects such as CUBRIDQA return HTTP 401 and are not supported, and any other Jira Server instance works by setting JIRA_URL. Atlassian Cloud (e.g. Hibernate HHH at hibernate.atlassian.net) is NOT supported as-is. It is the reverse of writing an issue up to JIRA. Triggers on phrases like 'CBRD-1234 내용 가져와', '이 이슈 md로 내려받아', 'jira 이슈 다운로드해서 참고', 'fetch jira issue as markdown'.
note
Record an exploration note (PoC, review, code analysis, or any work without a JIRA issue) as a Markdown file in your public work-docs repo. Give it a free-form category and a slug and it scaffolds <category>/<YYYY-MM-DD>-<slug>.md from an exploration template (목적/배경/범위·방법/발견/결론/다음 단계/참고), fills it from the session, and commits it. While researching, actively use the collection's lookup/analysis skills (cubrid-manual, cmt-manual, and Understand-Anything if installed) so findings are verified, not asserted from memory. Complements the worklog skill, which is for issue-keyed work. The repo is PUBLIC, so keep internal-only detail in the master DOCX. Triggers on phrases like 'PoC 노트 남겨', '검토 노트 작성', '코드 분석 기록', 'record an exploration note', 'note this analysis'.
report
Generate a CUBRID-house-style Korean analysis report as a Word (.docx): error analyses, code analyses, issue write-ups, before/after comparisons, and status/investigation reports. Builds the .docx with docx-js (the engine behind Anthropic's docx skill) from a JSON spec, embedding matplotlib charts, reproducing the exact design: centered cover (navy 23pt title), auto table of contents, navy/blue numbered headings, and bordered tables with row-level color coding (sky-blue header, green pass rows, red fail rows). Use when the user wants a Word report or structured document summarizing analysis, findings, comparisons, or current status of Hibernate/JDBC/CUBRID work. Triggers on phrases like 'write a report', '보고서 만들어', 'Word 문서로 정리', 'docx로 작성', 'analysis report', '검증 보고서', '비교 분석 문서'.
skill-create
Create a new skill in this claude-skills collection, from scaffolding through a lightweight trigger/dry-run test before commit. Use when the user wants to turn a workflow they just did into a reusable skill, capture a repeated process, or add a new skill to their toolbox. Applies this collection's conventions (domain prefixes, description-with-triggers, just install) plus skill-authoring practices from Anthropic's skill-creator (progressive disclosure, trigger-tested descriptions, bundled scripts/references/assets). Triggers on phrases like 'make this a skill', 'create a skill for this', 'save this as a skill', 'add a new skill', 'turn this workflow into a skill'.
worklog
Record a piece of issue work as a Markdown note in your public work-docs repo, for a shareable, PR-linkable trail (the GitHub-visible record that complements the internal master DOCX). Give it a CUBRID Jira key (CBRD, TOOLS, APIS, CUBRIDQA, ...) or a Hibernate HHH key (or a topic slug) and it scaffolds <KEY>/<KEY>-<slug>.md from a consistent template (배경/원인/변경/검증/결과/링크), fills it from the session's work, and commits it. Pairs with jira-fetch for issue context. The repo is PUBLIC: keep internal-only detail in the master DOCX. Triggers on phrases like '작업 기록 남겨', 'worklog 작성', '이 이슈 문서로 정리해서 커밋', 'record this issue's work as markdown', 'publish a work note for CBRD-1234'.
review-triage
Triage the review comments on your pull request. Fetches a PR's reviews (inline code comments, review summaries, and PR-level comments, from humans and bots) and, for each, gives an easy Korean summary, then judges it on the evidence (the actual code, the manual/spec, and this PR's own scope) instead of on the reviewer's authority, and either drafts a fix plan (valid) or a reasoned reply saying why not (wrong, already handled, out of scope, or style). Analysis and drafts only: it does not post replies or change code. Use when your PR has review feedback you want to understand and respond to. Triggers on phrases like 'PR 리뷰 정리해줘', '리뷰 코멘트 타당한지 봐줘', '리뷰 대응', 'triage my PR reviews', 'help me respond to PR review'.
pr-draft
Draft a pull request (title + body) from the current branch's commits and diff. The title is [XXX-0000] plus a concise, easy-to-understand English summary (the key is taken from the branch name); the body is Korean in three sections: Purpose (required, why this PR exists), Implementation (optional, how it was built), Remarks (optional, notes). Optional sections become N/A when empty. Passing `--tc` switches to a compact TC template for CMT (cubrid-migration) unit-test PRs only, summarized in a few lines with real counts (@Nested groups, tests run, `// DEFECT:` markers, suite before -> after); it is never selected automatically. Produces a copy-paste draft only; it does not create the PR. Triggers on phrases like 'PR 초안 작성', 'pr-draft', '이 브랜치 PR로 정리', 'draft a PR', 'PR 제목이랑 본문 만들어줘'.
blog
Draft a Korean IT/tech blog post for velog, minimal prose and visuals-first. Give it a slug (and topic) and it scaffolds work-docs/blog/<date>-<slug>/index.md, writes a tight Korean post (요약/들어가며/핵심/정리/참고), and renders every concept as an IMAGE because velog does NOT render Mermaid: diagram-as-code via Kroki, data charts via matplotlib, optional Excalidraw. Grounded in real code and manuals. Saves the draft to the public work-docs repo; you publish on velog. Triggers on phrases like '블로그 초안 작성', 'velog 글 써줘', '이거 블로그로 정리', 'write a tech blog draft', 'blog post about'.
submodule-bump
Pin a CUBRID submodule to a specific commit by dispatching the parent repo's 'Submodule bump (receiver)' GitHub Action, which opens or updates the bump PR on CUBRID/cubrid. Takes a submodule (cubrid-jdbc, cubrid-cci, cubridmanager: the short forms jdbc, cci, cms and the repo name cubrid-manager-server also work) and a commit SHA (or 'latest' for that submodule's develop head). Validates the submodule, the SHA, and the direction of the move before anything is dispatched, and shows the exact command for confirmation first. `--status` is a read-only overview: with no submodule it reports all three (pinned, head, how many commits behind, the open bump PR), and with one it lists that submodule's pending commits so you can pick how far to bump. Use when the automatic bump is stuck, when several submodule commits piled up and you want to catch up in one go, or when you need to pin one specific commit. Triggers on phrases like '서브모듈 상태 확인', '밀린 서브모듈 있나', '서브모듈 SHA 반영', 'jdbc 최신으로 올려줘', 'cci SHA 반영해줘', 'submodule bump
ci-check
Check a CUBRID PR's CI and, when a test job failed, work out whether it is your change, a project-wide failure, or a flaky test. Reads the PR's checks, pulls the failing tests from CircleCI, and classifies each one against recent runs of the same job on other PRs: 공통 (fails elsewhere too, not yours), flaky (result flips between runs), 고유 (fails only here). Also reports how far the PR's base is behind develop, and shows the failure output for the PR-specific ones. Answers 're-run, update the base, or actually investigate?'. Analysis only: it does not re-run CI or push anything. Triggers on phrases like 'CI 왜 깨졌어', 'CI 실패 원인 확인', '이 PR CI 확인해줘', 'flaky인지 봐줘', '재실행하면 되는지 확인', 'why did CI fail on this PR'.
weekly-report
Turn a list of JIRA issue keys into one short Korean line per issue, ready to paste into a weekly-report spreadsheet (one line per cell). Give it several keys (e.g. APIS-1111, TOOLS-4324) and it fetches each issue with the jira-fetch skill, reads the description, and compresses it into a single 개조식 line ending in 명사형: 대상 + 무엇을 하는가, no procedures or checklists. Summarizes what the issue IS, not progress or status. Use when writing the weekly report and you need each issue boiled down to one line. Triggers on phrases like '주간보고 정리', '주간 보고용으로 요약', '이 이슈들 한 줄로 정리', '주간보고 한 줄씩 만들어줘', 'weekly report lines for these issues'.
tutor
Manual, /tutor-only deep-learning tutor. Invoke ONLY when the user explicitly runs /tutor; do NOT trigger automatically (it is verbose by design and would waste tokens if auto-fired). Acts as a patient expert teacher for the current work or a named topic, starting from zero assumed terminology: it defines every term first, then teaches the concept step by step (big picture, core concepts, details with real examples, summary), grounded in the actual code and manuals, and ends with comprehension checks. Korean, and adaptive to follow-ups like 'easier / deeper / again / more examples'.
pr-review
Review SOMEONE ELSE'S pull request in depth, from a PR URL (or number) plus an optional Jira issue key. Fetches the whole repository locally as a detached git worktree (never touching your own branches or checkout), reads every changed file in full with no sampling, and establishes why the code is the way it is (git log, git blame, the PR body, the linked issue, prior PRs on the same files) before calling anything a defect: whatever it cannot ground is asked as a question to the author instead. Fans thirteen review passes out to parallel subagents, then merges them into one ranked Korean report with file:line evidence, split into 결함 / 판단 사항 / 범위 밖 / 질문. It reports to you and posts NOTHING: no review, no comment, no code edit. To judge review comments left on your own PR, use review-triage instead; to check your own unpushed work, use self-review. Triggers on phrases like '이 PR 리뷰해줘', 'PR 코드 리뷰 부탁해', '남의 PR 검토해줘', '이 풀리퀘 꼼꼼히 봐줘', 'review this PR', 'review someone else's pull request', 'deep review of PR 1629'.
self-review
Review your OWN in-progress change as a hostile stranger would, before any PR exists: run it mid-work or right before pushing. Gathers the contract first (Jira issue AC, commit messages, the diff's own claim, sibling conventions), then runs nine passes (contract conformance, completeness, premise verification, semantic delta, failure paths, blind spots, blast radius, convention, naming) and fans them out to parallel subagents so each angle is judged independently, then merges them into one ranked Korean finding list with file:line evidence. Its job is to counteract author tunnel vision by hunting the cases you never considered, including uncommitted and untracked work. Local analysis only: it never edits, commits, or posts. For judging other people's review comments on an existing PR, use review-triage instead. Triggers on phrases like '내 코드 셀프 리뷰', '푸시 전에 검토해줘', '작업 중인 변경 리뷰해줘', '빠뜨린 케이스 없는지 봐줘', '커밋 전에 스스로 리뷰', 'self-review my changes', 'review my own work before I push', 'what did I miss in this change'.
cmt-manual
Look up CUBRID Migration Toolkit (CMT) behavior in the official online manual at www.cubrid.org/cmt_manual: source database support (Oracle, MySQL, MariaDB, MSSQL, Informix, Tibero, and more), data type mapping, migration wizard steps, console/CLI mode, reports, and configuration. Use when migrating a database to CUBRID and you need to verify what CMT supports, how a source type maps to CUBRID, or how a CMT option or console command works. For CUBRID engine SQL/JDBC questions, use the cubrid-manual skill instead. Triggers on phrases like 'CMT', 'CUBRID Migration Toolkit', 'migrate X to CUBRID', 'CMT type mapping', 'how does CMT handle', 'CMT console mode'.
java-format
Apply google-java-format (AOSP style, in place) to Java files: the changed files, the staged files, all files, or specific paths. It asks which google-java-format version to use (e.g. 1.7 or 1.32.0), resolves the jar (GJF_JAR, then $GJF_DIR if set, then a cache, else auto-downloads that version), and runs it. Use when you want to format Java code to the project style, before committing. Triggers on phrases like 'google-java-format 적용', '자바 코드 스타일 적용', '변경된 자바 포맷', 'format the java files', 'apply java formatting'.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.