Blog
Latest articles
6 min readHow to Set Up MCP Servers in Claude Code and Cursor
Add MCP servers to Claude Code with one CLI command or to Cursor via mcp.json — complete 2026 setup for GitHub, filesystem, and remote HTTP servers.
4 min readBest MCP Servers for Developers in 2026
The MCP servers developers actually use in 2026 — GitHub, Playwright, databases, Figma, observability — ranked by workflow impact and context cost. Install three to five max.
4 min readWhy Does MCP Use So Many Tokens? (And How to Fix It)
MCP tool definitions can eat half your context window before you prompt. Here is why — and six fixes that actually work in Claude Code and Cursor.
6 min readHow Do Claude Code Skills Work? Complete SKILL.md Guide
Skills are SKILL.md playbooks Claude Code loads on demand — encode your team workflows once, trigger by slash command or context match. No more copy-pasting deploy checklists.
2 min readSkills vs MCP: When to Use Which in 2026
Skills teach the agent your workflow; MCP connects live systems. Here is when to use each — and the hybrid pattern HN devs settled on in 2026.
2 min readClaude Code vs Cursor: Token Cost and Efficiency (2026)
Independent 2026 benchmarks show Claude Code completing the same coding task in ~33K tokens vs Cursor ~188K — here is the dollar math and when each tool still wins.
5 min readOpenCode vs Claude Code: Open-Source Agentic CLI Compared
OpenCode runs Claude, GPT, Gemini, or local Ollama models in one terminal agent — Claude Code is official, polished, and Anthropic-native. Honest 2026 comparison.
5 min readHow to Build an MCP Server in TypeScript (Step-by-Step)
Build your first MCP server in TypeScript with the official SDK — tools, stdio transport, and a Claude Code test loop in under an hour.
12 min readClaude Opus 4.8 vs GPT-5.5 vs Gemini 3.1 Pro: June 2026 Benchmarks and Pricing
AA Index 61 vs 60 vs 57. SWE-Bench Pro, GDPval-AA, pricing tables, and where each model loses. Updated June 3, 2026 with primary source benchmarks.
11 min readDeepSeek V4 Pro and Kimi K2.6 vs Claude Opus 4.8: Open Weights at Frontier Level
MIT vs Modified MIT licenses, AA Index 52-54 vs 61, H100 self-host break-even math, and when open weights beat closed APIs. June 2026 guide.
11 min readHow to Use AI Models as Tools: Task Routing Matrix for Developers
Task-by-task picks: Opus 4.8 for refactors, GPT-5.5 for terminal agents, Gemini for RAG, DeepSeek V4-Flash for batch jobs. Printable routing table.
14 min readLLM Token Optimization in 2026: Model Routing, Caching, and Tool Budgets
100-turn coding session: $31 on Opus-only vs $2.65 routed. Six strategies with dollar math: caching, Sonnet default, MCP audit, structured output.