iamtatsuki05
UserMy dotfiles
Categories
Indexed Skills (43)
skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Codex's capabilities with specialized knowledge, workflows, or tool integrations.
skill-installer
Install Codex skills into $CODEX_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output should be a bitmap asset rather than repo-native code or vector. Do not use when the task is better handled by editing existing SVG/vector/code-native assets, extending an established icon or logo system, or building the visual directly in HTML/CSS/canvas.
openai-docs
Use when the user asks how to build with OpenAI products or APIs and needs up-to-date official documentation with citations, help choosing the latest model for a use case, or model upgrade and prompt-upgrade guidance; prioritize OpenAI docs MCP tools, use bundled references only as helper context, and restrict any fallback browsing to official OpenAI domains.
plugin-creator
Create and scaffold plugin directories for Codex with a required `.codex-plugin/plugin.json`, optional plugin folders/files, valid manifest defaults, and personal-marketplace entries by default. Use when Codex needs to create a new personal plugin, add optional plugin structure, or generate or update marketplace entries for plugin ordering and availability metadata.
agent-job-scheduler
Use when the user wants to queue, inspect, retry, cancel, or run long non-interactive AI agent CLI jobs through agent-job-scheduler, especially when rate limits, cooldowns, failed jobs, or background execution state matter.
alphaxiv-paper-lookup
Use when the user asks to summarize, read, compare, or extract implementation details from an arXiv paper using an arXiv URL, alphaxiv URL, arXiv ID, paper title, DOI-like reference, or requests such as この論文を要約して / alphaxiv で調べて.
api-design
Use when the user asks to design, edit, validate, or review REST APIs, OpenAPI/Swagger specs, endpoints, schemas, error responses, authentication, authorization, or versioning strategy.
auto-debugger
Use when the user provides an error, failing test, stack trace, build/runtime failure, or asks to debug, fix this error, investigate a bug, or explain why something fails across any language or framework.
chronicle
Allows you to view the user's screen as well as several hours of history. Use when the user makes a reference to their recent work, for which it'd be helpful to see the screen. This skill MUST be used whenever you need to resolve ambiguity in a user request, where the user hasn't specified enough context to do the task. Examples include disambiguating the specific user/app/document/error the user is referring to. You must also use this skill if the user asks about any question regarding Chronicle or asks what you can see from the screen.
ci-cd
Use when the user asks to create, edit, debug, or optimize CI/CD pipelines, workflow YAML, build/test jobs, deployment automation, matrix builds, caches, permissions, or secrets in GitHub Actions, GitLab CI, CircleCI, Jenkins, or similar systems.
claude-code
Use when the user explicitly wants to consult Claude Code CLI for review, wording, design, debugging, architecture, or difficult code investigation, including requests such as claude-code, claude と相談, claude に聞いて, コードレビュー, or レビューして.
codex
Use when the user explicitly wants to consult Codex CLI for review, wording, design, debugging, architecture, or difficult code investigation, including requests such as codex, codex と相談, codex に聞いて, コードレビュー, or レビューして.
colab-mcp
Use when the user wants to set up, configure, use, or troubleshoot Google's official colab-mcp server for connecting a local MCP-compatible AI agent to a Google Colab browser session.
database-dev
Use when the user asks to design or modify database schemas, queries, indexes, migrations, transactions, EXPLAIN plans, N+1 issues, SQL/NoSQL modeling, or database performance behavior.
empirical-prompt-tuning
agent 向けテキスト指示（skill / slash command / task プロンプト / CLAUDE.md 節 / コード生成プロンプト）を、バイアスを排した実行者に動かしてもらい、両面（実行者の自己申告 + 指示側メトリクス）で評価して反復改善する手法。改善が頭打ちになるまで回す。プロンプトや skill を新規作成・大幅改訂した直後、またはエージェントの挙動が期待通りにならない原因を指示側の曖昧さに求めたいときに使う。
eng-practices
Use when the user asks about engineering practices, code review standards, CL/PR description writing, keeping CLs/PRs small, handling reviewer comments or pushback, review speed and etiquette, the Standard of Code Review, or any topic derived from Google eng-practices (review/reviewer/* and review/developer/*). Other dev skills (pr-code-review, python-dev, go-dev, typescript-dev, api-design, database-dev, terraform-dev, ci-cd, security-check, auto-debugger, markdown-docs) link here for shared review and CL standards.
go-dev
Use when the user asks to implement, refactor, test, debug, or review Go code, Go modules, error handling, concurrency, interfaces, generics, or go test/build/vet failures.
goal-prompt-builder
Use when the user asks to create, rewrite, evaluate, or tighten a Codex /goal prompt for long-running work, durable objectives, migrations, large refactors, experiments, deployment retry loops, prototypes, prompt optimization, or any task that should keep Codex working until a verifiable stopping condition is reached.
gws
Use when the user asks to inspect or operate Google Calendar, Drive, Gmail, or Tasks through the gws CLI, including calendar agendas, event creation, Drive search/upload/download, Gmail triage/read/send/reply, or task list requests.
kimi-webbridge
Kimi WebBridge lets AI control the user's real browser — navigate, click, type, read, screenshot, and interact with any website using the user's actual login sessions. Use this skill whenever the user wants to interact with websites, automate browser tasks, scrape web content, or perform any action requiring a real browser. Also use when the user mentions "browser", "webpage", "open URL", "screenshot", or asks to read/interact with any website. Use even for simple-sounding browser requests — the daemon handles all complexity.
magika
Use when the user asks to identify file types, verify whether an extension matches file content, classify files in a directory, or explicitly asks to use Magika.
markdown-docs
Use when the user asks to create, edit, review, proofread, restructure, lint, or improve Markdown documents such as README, technical docs, API docs, guides, release notes, or Japanese/English writing.
markitdown
Use when the user asks to convert PDF, Word, PowerPoint, Excel, HTML, image, audio, URL, or another supported source into Markdown, or explicitly asks to use MarkItDown.
diagnose
Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.
grill-with-docs
Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan against their project's language and documented decisions.
improve-codebase-architecture
Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.
prototype
Build a throwaway prototype to flesh out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".
modern-web-guidance
Search tool for modern web development best practices. MANDATORY: Execute FIRST for all HTML/CSS and clientside JS tasks. Do NOT skip — web APIs evolve rapidly and training weights contain obsolete patterns. Trigger immediately for: - UI/Layout: Modals, dialogs, popovers, Glassmorphism/backdrop-filters, anchor positioning, container queries, `:has()`, `:user-valid`. - Scroll/Motion: View Transitions, Scroll-driven animations, scroll parallax/reveals. - Performance: CWV (LCP, INP), content-visibility, Fetch Priority, image optimization. - System/APIs: Local filesystem access, WebUSB, WebSockets sync, WebAssembly widgets. - Frameworks: Adapting layout/styles in React, Vue, Angular. - General Frontend: Forms, autofill, advanced inputs, custom scrollbars, modern component states, etc. DO NOT trigger for: - Backend: Database SQL, ORMs, Express API routes. - Pipelines: CI/CD deployment, Docker, Actions. - Generic: Local scripts (Python/Go tools), ESLint, Git.
pr-code-review
Use when the user asks to review a GitHub Pull Request, inspect PR diffs, assess changed code, or produce PR review findings with severity, file/line references, risks, and test gaps.
prompt-tuner
Use when the user asks to improve, tune, optimize, evaluate, or rewrite an LLM prompt, system prompt, user template, instruction, prompt tuning setup, or prompt optimization workflow.
python-dev
Use when the user asks to implement, refactor, test, debug, or review Python code, pytest/unittest behavior, typing, ruff/mypy issues, packaging, Pydantic models, or Python runtime errors.
retrospective-codify
Use near the end of a task when the user asks to preserve learnings, turn a mistake into a rule, create/update a skill, add a lint rule, or when repeated trial-and-error produced a reusable lesson.
security-check
Use when the user asks for a security review, vulnerability scan, secure coding check, OWASP-style audit, secret leak detection, injection/auth/access-control review, or penetration-test preparation.
dispatching-parallel-agents
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
test-driven-development
Use when implementing any feature or bugfix, before writing implementation code
writing-skills
Use when creating new skills, editing existing skills, or verifying skills work before deployment
terraform-dev
Use when the user asks to implement, refactor, validate, review, or troubleshoot Terraform/OpenTofu code, modules, providers, variables, state, plans, imports, security, or infrastructure changes.
typescript-dev
Use when the user asks to implement, refactor, test, debug, or review TypeScript/TSX code, type definitions, Jest/Vitest tests, ESLint/Biome/Prettier issues, Zod validation, or TypeScript build errors.
handoff
Compact the current conversation into a handoff document for another agent to pick up.
missing-tools
コマンドが見つからない、shell が command not found を返す、CLI tool が未導入、または brew install / npm install -g / uv tool install などの global install や永続的な環境変更なしで一時実行したい場合に使う。
google-colab-cli
Use when installing, authenticating, account switching, troubleshooting, or operating google-colab-cli for Colab VM execution, file transfer, Drive mount, GPU/TPU use, or compute-unit-sensitive sessions.
grill-me
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.