skills-by-yigitkonur

Use skill if you are exhaustively testing or release-gating martool CLI commands in a source checkout or deployed Coolify container over SSH, without local Docker or provider spend.

Use if driving agent-browser for Chrome/CDP automation, @ref snapshots, tabs, or verification.

Use if testing or debugging an iOS app via agent-device CLI — simulator flows, evidence, bug triage.

Use if supervising Jean agents through MCP and Computer Use for monitoring, recovery, or closure.

Use if auditing or designing a CLI for agent/LLM use — JSON output, exit codes, non-interactive.

Use if auditing or designing an MCP server for agent-readiness — framework, security, context.

Use if verifying claimed-done work or auditing session/plan/branch completion with evidence.

Use if hardening a SKILL.md by running a fresh subagent on a real task and fixing where it snags.

Use if auditing a running web app UI across pages/viewports, saving per-bug findings to a tree.

Use if auditing a running app's usability via persona journeys, saving per-issue findings to a tree.

Use if building or auditing UI against the 30 Laws of UX (Fitts, Hick, Gestalt, cognitive load).

Use if building or debugging a Chrome MV3 extension — manifest v3, service_worker, content_scripts.

Use if structuring or auditing TypeScript mcp-use/server code for Clean Architecture boundaries.

Use if sending email via Cloudflare Email Service, replacing Resend/SES/Postmark or send_email.

Use if building TypeScript with Effect-TS v3 — Effect.gen, Layer, Schema, typed errors, Stream.

Use if building browser-automation apps on the Kernel TS SDK (@onkernel/sdk) — browsers, pools.

Use if building TypeScript apps with langchain/@langchain — agents, RAG, structured output.

Use if integrating the LicenseSeat Swift SDK into a macOS/Swift app — activation, validation, seats.

Use if building a TypeScript MCP server on @modelcontextprotocol/sdk v1.x — single-package, Zod.

Use if building MCP servers on @modelcontextprotocol/server v2 alpha — split packages, registerTool.

Use if building TypeScript mcp-use MCPAgent code where an LLM orchestrates MCP tools (run/stream).

Use if writing TypeScript mcp-use MCP client code — MCPClient, MCPSession, useMcp, mcp-use/react.

Use if building TypeScript MCP servers with mcp-use/server — server.tool, ctx.auth, transports.

Use if authoring or fixing a Raycast Script Command (@raycast.* metadata header) — fields, modes.

Use if adding or auditing Sentry crash reporting in a macOS/Swift app — dSYM, breadcrumbs, tracing.

Use if creating, redesigning, or merging a Claude skill, with research before writing SKILL.md.

Use if building a TinaCMS + Next.js App Router site — tina/config.ts, MDX content, useTina editing.

Use if porting an MCP TypeScript server from @modelcontextprotocol/sdk v1.x to the v2 SDK.

Use if rebuilding a live URL or .html snapshot as a pixel-faithful AS-IS Next.js project.

Use if chasing a reproducible runtime bug or repeated failed fixes with a 4-phase root-cause pass.

Use if deploying/updating a docker-compose service on Coolify Cloud via its API — domains, env.

Use if creating, auditing, or migrating CLAUDE.md/AGENTS.md/REVIEW.md instruction files.

Use if onboarding a repo to Jean — jean.json and .worktreeinclude setup, run, teardown, ports.

Use if scaffolding Makefile targets for dev, tunnels, deploys, R2, Supabase, Railway, Vercel.

Use if controlling or testing a connected Android phone via the mobilerun CLI — tap, type, swipe.

Use if publishing to npm via GitHub Actions — trusted publishing, provenance, semantic-release.

Use if driving a large ambiguous initiative with scored question alignment before parallel exec.

Use if fanning out parallel Codex adversarial reviews, verifying findings, fixing in worktrees.

Use if running Codex review loops across branches, lenses, comparisons, or convergence rounds.

Use if running deep multi-file research over 5+ entities or a market — wave-dispatched corpus.