Yigit Konur

47 skills for AI coding agents.

Building a SwiftUI app with Convex backend and Clerk auth — covers reactive subscriptions, TypeScript server functions, schema design, auth integration, and iOS/macOS production patterns

Creating or refining AI code review config for GitHub Copilot, Devin Bug Catcher, Greptile, or any combination for a repository

Writing, reviewing, or setting up Swift code quality tooling for macOS/iOS apps — covers SwiftLint, SwiftFormat, Airbnb rules, architecture, testing, and platform-specific patterns

Building or debugging a Chrome extension with Manifest V3 — service workers, content scripts, popup/sidepanel, messaging, storage, permissions, testing, or Web Store publishing

Building TypeScript applications with the GitHub Copilot SDK (@github/copilot-sdk), including sessions, tools, streaming, hooks, custom agents, or BYOK

Building UI with daisyUI components, converting Figma designs or screenshots to daisyUI code, or generating themes using the daisyui-blueprint MCP server

Building applications or automation systems on hcom — multi-agent architectures, custom workflow scripts, CI/CD integration, or understanding hcom internals like hooks, SQLite, relay, and events

Building TypeScript agents, RAG pipelines, or tool-calling workflows with LangChain.js and LangGraph

Building TypeScript AI agents with mcp-use MCPAgent — configuration, LLM integration, streaming, structured output, observability, and production deployment

Building or fixing an mcp-use server that renders React widgets and needs useWidget, useCallTool, streaming, CSP, or ChatGPT/MCP Apps compatibility

Building MCP client applications with the mcp-use TypeScript SDK — MCPClient, sessions, tools, resources, React hooks, code mode, and authentication

Building or extending TypeScript MCP servers with the mcp-use library — tools, Zod schemas, resources, prompts, transports, OAuth, sessions, testing, and deployment

Building OpenClaw plugins — registering tools, channels, model providers, writing plugin manifests, and configuring tool profiles

Creating, testing, or publishing OpenClaw skills and need SKILL.md structure, metadata gating, reference routing, or ClawHub guidance

Building OpenClaw automation — Lobster typed workflows, cron scheduled jobs, LLM task chains, and browser automation

Building or repairing Raycast Script Commands in Python or Bash and need correct metadata, output modes, arguments, discovery, or runtime behavior

Creating or substantially revising a Claude skill and need workspace-first evidence, remote comparison, and repo-fit synthesis before writing SKILL.md

Extending a Supastarter app with App Router pages, oRPC procedures, Better Auth flows, Prisma/Drizzle data work, billing, storage, or monorepo package changes

Converting live websites or saved HTML snapshots into grounded Next.js pages with browser capture, self-hosted assets, extracted styles, and screenshot-led fidelity verification

Converting Vue 2/3 or Nuxt components to Next.js App Router with React, including component mapping, state migration, routing, SSR, or ecosystem packages

Debugging a Tauri app with CrabNebula DevTools — inspecting Rust logs, IPC calls, runtime config, plugin permissions, or bundled assets

Applying Clean Architecture layers, dependency direction, DDD entities, use cases, or strict TypeScript patterns to structure or audit a codebase

Building, modernizing, or reviewing macOS SwiftUI code with Apple's Liquid Glass design system — glass toolbars, sidebars, windows, NSGlassEffectView, or migrating from NSVisualEffectView

Building Fastify REST APIs with TypeBox schemas, type-provider-typebox, route validation, OpenAPI generation, plugins, hooks, inject() testing, or Drizzle/Prisma database integration

Writing or reviewing framework-agnostic TypeScript and need strict typing, tsconfig/lint decisions, safer refactors, or guidance on generics, unions, and typed boundaries

Enhancing a prompt before sending it to an LLM or coding agent — adds narrative structure, thinking-steering, failure pre-emption, halt conditions, and verification layers

Improving a skill's instructional quality by having a subagent use it on a real task, reading the execution trace to find breaks, and applying fixes

Forensically documenting the visual system of an existing SaaS dashboard or admin UI into structured design docs, not rebuilding or redesigning it

Generating, auditing, or migrating repo-grounded AGENTS.md and CLAUDE.md files, especially for dual-file setups, multi-agent compatibility, or converting existing agent config

Creating or refining GitHub Copilot code review instruction files such as .github/copilot-instructions.md and scoped *.instructions.md for repository-specific PR reviews

Creating or refining REVIEW.md, AGENTS.md, or repo-scoped Devin Bug Catcher review instructions

Generating `.greptile/` review config, scoped rules, or context-file mappings for Greptile pull request reviews in a specific repository

Configuring an OpenClaw agent workspace or runtime policy: workspace files, tool profiles, allow/deny lists, skills loading, memory, or security settings

Optimizing, auditing, or enhancing an MCP server's tool quality, error handling, security, context efficiency, or production readiness for any MCP SDK

Planning a project as GitHub Issues with sub-issues, wave-based execution order, labels, and dependency wiring via the gh CLI

Writing a PRD, product requirements document, feature spec, or requirements document with structured discovery, codebase-grounded analysis, and AI-agent-actionable output

Framing a vague request, comparing approaches, prioritizing a backlog, or defining risks, checkpoints, and decision criteria before acting

Automating npm publishing via GitHub Actions and need auth, versioning, provenance, or workflow-template choices

Reviewing a GitHub pull request or branch diff for merge readiness with high-signal findings grounded in stated goals and changed-file clusters

Driving the agent-browser CLI for browser navigation, DOM-grounded interaction, session/tab management, screenshots, data extraction, or repeatable web app workflows

Orchestrating multi-domain research across official docs, community forums, and practitioner experience with parallel researcher agents producing structured, implementation-ready documentation

Evaluating GitHub repositories for quality, maturity, and maintainer credibility using GraphQL batch queries, REST signals, and code-level analysis

Searching GitHub for repositories matching a technology, pattern, library, or framework using diverse search hypotheses and gh CLI

Finding the best GitHub repos for a technology need: searches, evaluates, and ranks repos using subagent swarms with automatic retry when results are weak

Orchestrating multiple AI coding agents via hcom — launching headless agents, messaging between them, or writing multi-agent workflow scripts with patterns like worker-reviewer loops

Executing a project plan stored as GitHub Issues with sub-issues by reading the tree, dispatching subagents wave-by-wave, and tracking completion

Orchestrating OpenClaw multi-agent systems — spawning sub-agents, managing sessions, ACP routing, and cross-agent messaging

Deploying OpenClaw to production — Docker containers, VPS setup, security hardening, gateway management, and monitoring

Controlling a live browser with @anthropic-ai/playwright-cli for navigation, forms, screenshots, tab workflows, or interactive debugging from the terminal

Researching a bug, library choice, architecture trade-off, or technical question where current web evidence, Reddit experience, and multi-source synthesis matter

Running SEO workflows with MCP Marketers tools — site audits, keyword research, backlink analysis, competitor intelligence, SERP analysis, rank tracking, or content optimization

Testing or debugging MCP servers with mcpc CLI across stdio, HTTP stateless, and HTTP stateful transports — tool calling, schema validation, and automated test scripts

Searching, discovering, and downloading AI coding skills using skill-dl with Serper-powered search and batch downloads