Signal, not noise
Long-form writing on Model Context Protocol, agentic AI tools, and the plumbing that actually makes developers productive.
MCP config hell: why every AI coding tool stores the same thing differently
Claude wants JSON with `mcpServers`. VS Code wants a `type` field. Zed nests under `context_servers`. Codex uses TOML. Continue uses YAML. Here's why that happened — and the cost every developer is paying.
Introducing MCPBolt — one config, every AI tool
After six months of copy-pasting the same Supabase server into eight different JSON files, we built the tool we wanted. Here's what MCPBolt does and why it stays local-only.
What is the Model Context Protocol? A developer's field guide
MCP is Anthropic's open protocol that lets AI tools plug into external data and actions. Here's a plain-English tour: what it is, how it differs from tool-use, and why every major IDE adopted it in under a year.
The top MCP servers to install in 2026
We surveyed the ecosystem: Supabase, Context7, PostHog, Playwright, GitHub, Linear, Sentry, and more. Here's what each one does, when you'd install it, and the one-click config for MCPBolt.
MCPBolt vs. manual editing vs. built-in tool UIs: a developer's comparison
You have three options for managing MCP servers: edit JSON by hand, use each app's built-in UI, or let MCPBolt do it. We benchmarked all three on install time, error rate, and cross-app sync.
Why MCPBolt stays local-first — and always will
No telemetry. No account. No cloud sync of your secrets. Here's the engineering reason your MCP tool should never phone home, and how MCPBolt enforces it at every layer.
The MCP ecosystem in 2026: market landscape
Who's building MCP servers, who's using them, and where the next wave of agentic tooling is heading. A state-of-the-market report based on GitHub stars, registry submissions, and developer surveys.