axross

Creating, refining, splitting, consolidating, renaming, or auditing an agent skill — drafting a `SKILL.md`, editing frontmatter, tightening a `description`, deciding where a new rule belongs, or running the structure validators. Triggers on "add a skill",…

Writing, refactoring, or reviewing code for maintainability or design — "readable", "too long", "refactor", "abstraction", "cohesion", "magic number", "dead code", "what should this be called", or "should this live elsewhere". Cohesion as the test of what…

Writing, reviewing, refactoring, or running end-to-end tests, or verifying a change through the e2e suite — including a prompt naming only snapshots, test IDs, selectors or locators, polling and waiting, flaky tests, scenario coverage, a runner (Playwright,…

Driving a code change or document update end-to-end through the plan → code → review loop — "deliver this issue", "implement and open a PR for X", a free-form change request, or resuming an in-progress run — as the project's default change loop. Apply even…

Reviewing whether a change carries adequate verification evidence — "is this verified", "did this break anything", "were the required checks run". The reviewer's QA pass on top of the development verification rules. Covers requiring command evidence for the…

Building or reviewing a React component — "props", "spread props", "compound component", "variant prop", "empty state", "error state", "skeleton", "data-testid", "testID", "extract a hook", "useEffect", "where state lives", "memo", "virtualize", "FlatList",…

Writing, reviewing, or refactoring the styles of a React component on web or mobile native — a CSS Module, a Unistyles `StyleSheet.create`, a theme or token file, or global styles. Triggers on "styling", "CSS", "theme", "design token", "dark mode", "container…

Every task that touches a project — implementing, refactoring, running a project command, or writing a pull request body — even when the request never mentions formatting, linting, comments, dependencies, docs, or descriptions. Also when you need to know how…

Drive one unit of work — a GitHub issue, a pull request, or a free-form prompt — from intake to a review-ready pull request in one continuing session; plan and pause for human approval, implement and verify, request the independent review, and address its…

The authoring discipline for agent skills under the host project's skill root (e.g., `.claude/skills/`) — agentskills.io format, frontmatter management and invocation-control fields, kebab-case naming, description and when_to_use writing for discovery,…

The security and privacy review lens for code changes. Covers secrets/env vars, the framework's public/client-exposed env-var prefix, input validation, public exposure, output-encoding/injection in rendered content, SSRF/outbound fetching, error-reporting…

The review methodology for every review task, including post-implementation self-review. Covers reviewer-mode reset, diff scoping with `git status`/`git diff`, severity labels, file-line evidence, fix snippets, verification evidence, second-pass re-review…

How to build React components in this project. Covers component anatomy and naming, CSS Modules pairing, Base UI headless primitives, server/client boundary decisions, test-id hooks, accessibility mechanics, and when a repeated pattern earns a shared…

Conventions for MDX content pages on this Pokémon TCG Pocket documentation site. Covers page placement and metadata exports, heading structure, tables and examples, internal/external linking, tone, game terminology, and the unofficial-fan-content boundary.

The project's baseline working rules. Covers the format/lint loop, scoped change management, current-docs lookup triggers, run-script commands, type-safety discipline, source-comment and doc-comment conventions, verification requirements, MDX content-layer…

Conventions for Playwright end-to-end tests in this project. Covers the test directory layout and file naming, structured test/step naming, stable test-id chained/scoped locators with a role-then-copy fallback hierarchy (text matching only when asserting the…

Drive one unit of work — a GitHub issue, a pull request, or a free-form prompt — from intake to a review-ready pull request in one continuing session; plan and pause for human approval, implement and verify, request the independent review, and address its…

The authoring discipline for agent skills under the host project's skill root (e.g., `.claude/skills/`) — agentskills.io format, frontmatter management and invocation-control fields, kebab-case naming, description and when_to_use writing for discovery,…

The security and privacy review lens for code changes. Covers secrets/env vars, Next.js's NEXT_PUBLIC_ client-exposed env-var prefix, input validation, public exposure, output-encoding/injection in rendered untrusted content, SSRF/outbound fetch of…

The review methodology for every review task, including post-implementation self-review. Covers reviewer-mode reset, diff scoping with `git status`/`git diff`, severity labels, file-line evidence, fix snippets, verification evidence, second-pass re-review…

Conventions for React components and hooks in this project — the shared-vs-feature tier boundary, promotion criteria for repeated UI, component anatomy (naming, props, server/client boundary, test-id hooks), CSS Modules styling composition, logic extraction…

The project's baseline working rules. Covers the format/lint loop, scoped change management, current-docs lookup triggers, run-script commands, type-safety discipline, source-comment and doc-comment conventions, verification requirements, end-to-end test…

Conventions for Playwright end-to-end tests in this project. Covers the test directory layout and file naming, structured test/step naming, stable test-id chained/scoped locators with a role-then-copy fallback hierarchy (text matching only when asserting the…

How an agent reads and writes GitHub (issues, pull requests, comments, labels, reviews, branches) through a proxy-mediated single-operator identity, as a Claude Code + GitHub MCP harness does. Covers routing everything through the one sanctioned tool channel,…

Creating, refining, splitting, consolidating, renaming, or auditing an agent skill — the authoring rules for the agentskills.io format, and the validators that enforce them. Triggers on drafting a `SKILL.md`, editing frontmatter, tightening a `description`,…

Deciding whether material belongs in a skill at all or in the project's own documentation, and then which tier it lives in — the storage, install, and drift model for a project keeping skills in two tiers. Triggers on adding, editing, renaming, moving, or…

Writing or reviewing code that handles untrusted input, secrets, outbound requests, rendered content, or third-party dependencies — the OWASP Top 10:2025 lens in two modes, writing secure-by-default code and judging the risk a change introduces. Triggers on…

Writing, refactoring, or reviewing code for maintainability or design — keeping what you change readable, cohesive, and cheap to change next time. Triggers on "readable", "too long", "refactor", "abstraction", "cohesion", "magic number", "dead code", "what…

Reviewing a code change — a pull request, a branch or commit-range diff, or a post-implementation self-review of your own work before calling it done. The methodology for judging whether a change already written is safe to merge, and for reporting why. Not…

Authoring or revising a Git commit message, titling a pull request, or judging whether a header conforms — the Conventional Commits v1.0.0 header contract, plus a validator that catches a malformed header before it reaches your history. Triggers on "write a…

Writing, reviewing, refactoring, or running end-to-end tests, or verifying a change through the e2e suite — whole-system checks that drive a running build the way a real user or client does. Triggers on snapshots, test IDs, selectors or locators, polling and…

Reading or writing GitHub from inside an agent session that shares one connected operator identity — any issue, pull request, comment, label, review, or branch operation, not only end-to-end change loops. The operating model that keeps those writes safe and…

Drive one unit of work — a GitHub issue, a pull request, or a free-form prompt — from intake to a review-ready pull request in one continuing session; plan and pause for human approval, implement and verify, request the independent review, and address its…

The authoring discipline for agent skills under the host project's skill root (e.g., `.claude/skills/`) — agentskills.io format, frontmatter management and invocation-control fields, kebab-case naming, description and when_to_use writing for discovery,…

The security and privacy review lens for code changes. Covers secrets/env vars, the framework's public/client-exposed env-var prefix (NEXT_PUBLIC_), input validation, access control, public exposure, output-encoding/injection in rendered untrusted content,…

The review methodology for every review task, including post-implementation self-review. Covers reviewer-mode reset, diff scoping with `git status`/`git diff`, severity labels, file-line evidence, fix snippets, verification evidence, second-pass re-review…

The project's baseline working rules. Covers the format/lint loop, scoped change management, current-docs lookup triggers, run-script commands, source-comment and TSDoc doc-comment conventions, verification requirements, data-layer/migration handling,…

Conventions for Playwright end-to-end tests in this project. Covers the `e2e/tests` layout and `<name>.test.ts` naming, `test.step` structuring, stable test-id chained/scoped locators (never text matching), framework-native auto-waiting assertions over manual…

How an agent reads and writes GitHub in this repository's cloud environment — issues, pull requests, comments, labels, reviews, or branches. Covers the sanctioned tool channel (the built-in `mcp__github__*` tools), operating as the connected operator…

Move one unit of in-progress work across session boundaries via a self-contained, downloadable handoff package — a handoff-<unix epoch>.md document plus an optional same-epoch .zip of patches and artifacts. Suspends the outgoing session's state at a safe…

Apply this skill at the start of EVERY review task, including post-implementation self-review. Covers reviewer-mode reset, diff scoping with `git status`/`git diff`, severity labels, file-line evidence, fix snippets, Conventional Commit and PR-title checks,…

Apply this skill at the start of EVERY task in this project. Covers workflow classification, scoped change management, Biome format/lint expectations, npm scripts, Conventional Commits, pull request titles for squash merge, TanStack…

Use this skill when reviewing security or privacy implications in the password generator. Covers local-only generation, Web Crypto, clipboard use, generated password exposure, storage and URL prohibitions, telemetry limits, environment variables,…

Use this skill whenever writing, reviewing, refactoring, or running Playwright end-to-end tests for the password generator. Covers `e2e/` placement, primary generator flow, desktop/mobile projects, layout and option changes, generated output assertions,…

Use this skill when adding, editing, testing, or reviewing keyboard layout data and layout-derived character sets. Covers QWERTY reference ownership, physical key positions, direct and shifted outputs, AltGr ambiguity, metadata, source notes, deterministic…

Use this skill when reviewing readability, naming, abstraction boundaries, complexity, dead code, or scope discipline. Covers pure password and layout modules, small React components, dependency restraint, clear naming, Biome complexity signals, avoiding…

Use this skill whenever writing, reviewing, or modifying code that throws, catches, reports, logs, or displays runtime failures in the password generator. Covers console output, error boundaries, browser API failure messages, generated-password secrecy,…

Use this skill when implementing, testing, reviewing, or changing password generation behavior. Covers QWERTY-selected-layout same-position character pools, Web Crypto randomness, unbiased index selection, entropy estimates, complete-word output, category…

Driving a code change or document update end-to-end through the plan → code → review loop — "deliver this issue", "implement and open a PR for X", a free-form change request, or resuming an in-progress run. The project's default change loop, carrying one unit…

Wiring or reviewing an Amplitude integration — the vendor layer, covering SDK init, identity and sessions, autocapture, tracking plans, consent, replay, ingestion, and cost. Triggers on `@amplitude/analytics-browser`, `@amplitude/unified`, `amplitude.init`,…

Working on an Expo app's framework layer — the shell around the components, covering routes, app config, config plugins, permissions, safe areas, assets, on-device storage, testing, SDK upgrades, and shipping a build or update. Triggers on "expo-router",…

Designing, building, or reviewing a high-fidelity user-facing surface with real colors, type, spacing, and states — the visual design system behind a production-ready surface, and the rationale behind each decision in it. Triggers on "visual design", "design…

Working on Jest itself — the runner layer rather than the tests' content, covering its config, API, CLI, and transform and isolation mechanics. Triggers on a Jest config, `jest.fn`, `jest.spyOn`, `jest.mock`, `unstable_mockModule`, `useFakeTimers`,…

Writing or reviewing a Next.js App Router change — the version 16 framework layer, covering routing, rendering, caching, mutations, and deployment. Triggers on "app router", "server component", "use client", "use cache", "server action", "route handler",…

Building or reviewing a React component — how one is put together, from composition and props contracts through extracted logic, state, memoization, and testability, on web and mobile native alike. Triggers on "props", "spread props", "compound component",…

Writing, reviewing, or refactoring the styles of a React component on web or mobile native — the implementation mechanics of tokens, responsiveness, adaptive states, and platform stylesheets. Triggers on a CSS Module, a Unistyles `StyleSheet.create`, a theme…

Driving a code change or document update end-to-end through the plan → code → review loop — "deliver this issue", "implement and open a PR for X", a free-form change request, or resuming an in-progress run — as the project's default change loop. Apply even…

Writing, reviewing, or refactoring the styles of a React component on web or mobile native — a CSS Module, a Unistyles `StyleSheet.create`, a theme or token file, or global styles. Triggers on "styling", "CSS", "theme", "design token", "dark mode", "container…

Writing, refactoring, or reviewing code for maintainability or design — "readable", "too long", "refactor", "abstraction", "cohesion", "magic number", "dead code", "what should this be called", or "should this live elsewhere". Cohesion as the test of what…

Writing, reviewing, refactoring, or running end-to-end tests, or verifying a change through the e2e suite — including a prompt naming only snapshots, test IDs, selectors or locators, polling and waiting, flaky tests, scenario coverage, a runner (Playwright,…

Working on an Expo app's framework layer — routes, app config, config plugins, permissions, safe areas, assets, on-device storage, testing, SDK upgrades, or shipping a build or update. Triggers on "expo-router", "prebuild", "CNG", "EXPO_PUBLIC_", "splash…

A change touching Jest itself — a Jest config, `testMatch`, `@jest/globals`, `jest.fn`, `jest.spyOn`, `jest.mock`, `unstable_mockModule`, `useFakeTimers`, `toMatchSnapshot`, `moduleNameMapper`, `transformIgnorePatterns`, `setupFilesAfterEnv`, `next/jest`,…

Reviewing whether a change carries adequate verification evidence — "is this verified", "did this break anything", "were the required checks run". The reviewer's QA pass on top of the development verification rules. Covers requiring command evidence for the…

Writing, reviewing, or modifying code that logs, throws, catches, reports an error, tracks a product event, identifies a user, or configures a logger, error tracker, tracing, or analytics tool — including a request naming only "logging", "log level", "capture…

OWASP-framed security and privacy rules for this small Hono URL proxy. Use when writing or reviewing code that handles proxy payloads, UUIDv5 short keys, KV short-link storage, Obsidian URI construction, source URLs, Open Graph metadata, environment…

Apply this skill whenever creating, refining, restructuring, renaming, or auditing an agent skill under the host project's skill root (e.g., `.agents/skills/`) - drafting a `SKILL.md`, splitting a skill into reference files, tightening a `description`, or…

Rules for reviewing diffs in this Hono URL proxy. Use when reviewing a pull request or local diff: covers review scope, severity, evidence and citation format, tone, escalation boundaries, and routing to security, QA, maintainability, observability,…

Rules for implementing, refactoring, and shipping code in this Hono URL proxy app. Use for any code change: covers current Hono/Sentry docs, npm command sequence, route/view/helper boundaries, import hygiene, dependency rules, verification, and commit message…

Use when writing, reviewing, running, or maintaining Playwright end-to-end tests in this Hono URL proxy app. Covers `playwright.config.ts`, `npm run test:e2e`, `PLAYWRIGHT_BASE_URL`, browser installation, `e2e/tests/routes/**` placement, `.test.ts` naming,…

Rules for keeping this small Hono URL proxy readable and easy to evolve. Use when writing or reviewing code for boundaries, pure helper placement, validation centralization, Hono JSX and CSS complexity, dead code, type reuse, kebab-case file naming, camelCase…

Rules for error handling, structured logging, and Sentry error tracking in this Hono URL proxy. Use when code decodes external payloads, handles expected invalid links, touches KV-backed bridge resolution, throws route/helper errors, logs runtime diagnostics,…

Navigation reference for this Hono URL proxy repo. Use when writing, reviewing, or navigating code: covers the Cloudflare Worker entry under src/common/worker.tsx, common/core source under src/common/, Obsidian route modules under src/obsidian/, colocated…

Apply this skill to process the Knowledge Base › Articles queue end-to-end — read every row with Status "New", write a structured 1–12 section summary into the row's page content, transition the row to "Summarized", synthesize wiki-style reference sub-pages…

Apply this skill whenever any agent reads from or writes to the Knowledge Base › Articles database in Notion (the dev-daily knowledge base of tech articles). Triggers include "save this article to Notion", "add to articles DB", "find unread articles", "mark…

Apply this skill when the user wants to find, retrieve, and store recent tech news, articles, or blog posts (English or Japanese / 日本語) within a publish-date range into the dev-daily Knowledge Base › Articles database in Notion. Triggers include "fetch…

Apply this skill whenever creating, refining, restructuring, renaming, or auditing an agent skill under the host project's skill root (e.g., `.agents/skills/`) — drafting a `SKILL.md`, splitting a skill into reference files, tightening a `description`, or…

The user's personal English writing voice — Kohei Asai (GitHub axross) — for text he publishes or sends under his own name across GitHub PR descriptions, PR and code-review comments and replies, commit messages, Confluence pages and engineering blog posts,…