skills

Configure rate limiting, manage auth secrets, set up CSRF protection, define trusted origins, secure sessions and cookies, encrypt OAuth tokens, track IP addresses, and implement audit logging for Better Auth. Use when users need to secure their auth setup, prevent brute force attacks, or harden a Better Auth deployment.

Create, critique, or revise prompts for Claude Design and similar AI UI prototype generators. Use when the user wants a visual UI prototype prompt, wants to improve an existing design prompt, shares a generated design for review, or needs prompts that preserve product semantics, domain rules, workflow state, and negative constraints instead of producing generic UI.

Write engineering design documents for architecture, API, framework, or product-contract decisions before implementation planning. Use when the user asks for a design doc, proposal, RFC-style design, contract design, or wants to debate system shape. Not for execution plans, milestone checklists, progress trackers, or implementation task breakdowns.

Write requested implementation plans, milestone checklists, execution trackers, and handoff plans from an accepted design or exact implementation direction. Use for planning artifacts only, not for ordinary implementation requests or design debate.

This skill should be used when the user asks to "review file organization", "check code layout", "identify layout smells", "audit folder structure", "review module organization", "check file naming", "find code structure issues", "review how code is organized", or mentions that AI-generated code is disorganized or hard to navigate. Works with Python and TypeScript codebases.

Share skills and top-level instructions across coding agents (Claude Code and others) via symlinks. Use for "set up multi-agent", "share skills between agents", "link AGENTS.md to CLAUDE.md", "unify agent configs", or pointing `.agents/skills` and `.claude/skills` at the same place. Auto-detects which side (`.agents/`+`AGENTS.md` or `.claude/`+`CLAUDE.md`) already has content and makes the other a symlink into it; defaults to `.agents/` canonical when starting fresh.

Set up Python quality gates (ruff lint/format, type checking, pytest + coverage, complexity, dead code, duplicates, file length) in any Python repo, wired through `prek` (pre-commit reimagined) with a `check.sh` orchestrator underneath. This skill should be used when the user says "add quality gates", "set up linting", "add python checks", "quality gate setup", "add prek", "add check script", "enforce code quality", or wants to establish code quality infrastructure in a Python project.

AlphaTab music notation in JS/TS apps. Use for @coderline/alphatab setup, Vite worker integration, fonts/soundfonts, playback controls, track rendering, AlphaTex, drum notation, and React integration.

Configure, scaffold, or troubleshoot Better Auth in TS/JS apps. Use for auth.ts, login/sign-up pages, database adapters, sessions, plugins, OAuth, email/password, email verification, password reset, 2FA/MFA/OTP/TOTP, backup codes, organizations/teams/RBAC, or migrating from another auth library.

Turn development session transcripts into blog-writing notes. Use to summarize a coding session, extract a narrative arc from Claude Code or Codex transcripts, combine sessions, or draft BLOG-NOTES.md for a technical post.

Audit Linear issues against the current codebase and close only those clearly done or no longer relevant. Use to triage backlog tickets, clean up stale backend issues, or reconcile Linear with recent implementation work.

Write robust, maintainable tests. Use when writing unit or integration tests, debugging test failures, or setting up fakes/mocks. Covers test design, assertions, and test doubles.

Review or refactor frontend file layout, naming, and feature boundaries. Use when code feels AI-organized, route/screen/container ownership is unclear, or feature barrels and shared UI boundaries need cleanup.

Refactor UI by collapsing distinctions without difference, naming components by responsibility (not schema entities), pushing data sourcing out of UI, and promoting generic primitives into shared UI. Use when a UI feels over-specialized, entity-named, or too smart about data fetching.

Review and refactor Go code for modern release features (Go 1.25 and 1.26) — new language and stdlib APIs, `go fix` workflows, tooling, and newly deprecated or tightened behaviors. Use after upgrading Go, during code review, or when modernizing concurrency, error handling, logging, testing, http/proxy, or reflection code.

Hunt for frontend issues in SQLite debug logs. Use for "hunt for issues", "check logs", "log hunt", "find bugs in logs", "debug session", triaging errors/warnings from the SQLite debug DB, or wiping the log database.

Token-efficient SQLite-based debug logging for frontend projects — local log server, structured logs, AI-friendly triage that persists across page refreshes. Use for "add debug logging", "set up SQLite logging", "AI-friendly logging", or frontend observability.

Author or edit `.stories.ts(x)`/`.mdx`, configure `.storybook/` (main/preview/manager), restructure sidebar/hierarchy, debug missing stories or single-story hoisting, write play functions, or use `composeStories` in Vitest/Jest/Playwright. Also triggers on `Meta<typeof ...>`, `StoryObj`, or `@storybook/*` imports. Scope: CSF3, React + TS, v10.3.

SurrealDB and SurrealQL — schema design, auth/permissions, deployment, SDKs, SurrealDB Cloud, Surrealist, integrations. Use when implementing or debugging SurrealDB-backed systems, writing SurrealQL, modeling data, or answering product/SDK questions.

Set up Go quality gates (build, vet, golangci-lint, goimports, duplicates, dead code, complexity, mod tidy drift, coverage) in any Go repo, wired through `prek` (pre-commit reimagined) with a `check.sh` orchestrator underneath. Use when the user says "add quality gates for Go", "set up Go linting", "add golangci-lint", "add go checks", "Go quality gate setup", "add prek for Go", or wants to establish code quality infrastructure in a Go project (as opposed to TypeScript or Python).

Set up Rust quality gates (cargo check/build, clippy, rustfmt, dead code, unused deps via cargo-machete, doc build, tests) in any Rust repo, wired through `prek` (pre-commit reimagined) with a `check.sh` orchestrator underneath. Use when the user says "add quality gates for Rust", "set up Rust linting", "add clippy", "add cargo clippy and rustfmt", "add cargo checks", "Rust quality gate setup", "add prek for Rust", "add cargo-machete", "Rust pre-commit hooks", or wants to establish code quality infrastructure in a Rust project (as opposed to Go, TypeScript, or Python). Step 12 covers optional cargo-audit/cargo-deny, coverage via cargo-llvm-cov, MSRV verification, miri, and feature-matrix testing via cargo-hack for larger repos.

Set up TypeScript quality gates (typecheck, ESLint, Prettier, complexity, dead code, duplicates, coverage, file length) in any TS repo, wired through `prek` (pre-commit reimagined) with a `check.sh` orchestrator underneath. Use when the user says "add quality gates", "set up linting", "add ts checks", "quality gate setup", "add prek", "add check script", or wants to establish code quality infrastructure in a TypeScript project.

Write and debug TypeQL queries for TypeDB 3.8+. Use when working with TypeDB schemas, data queries, insertions, deletions, or functions. Covers schema definition, CRUD operations, pattern matching, aggregations, and common pitfalls.

A practical, task-oriented implementation of Martin Fowler's refactoring approach, used when improving design without changing behavior.

Use the type-bridge Python ORM for TypeDB. Covers defining entities, relations, attributes, CRUD operations, queries, expressions, and schema management. Use when working with TypeDB in Python projects.