Eggthropic is an independent experimental project and is not affiliated with, endorsed by, or officially connected to Anthropic.

Notes

Anthropic updates, decoded

Short, practical articles on Claude and Anthropic developments. Each note covers what changed, why it matters to developers, what can be built with it, and what limitations or risks to watch for.

Claude Code

Claude Code: What Agentic Coding Looks Like in Practice

Claude Code moves beyond autocomplete — it reads your full codebase, runs commands, and ships committed code across multiple files. Here's what that actually means for daily developer workflows.

What can be built: Automated PR description generation from git diff (see our Agent Skills experiment) + 4 more

Agent Skills

Agent Skills: Giving Claude Persistent, Portable Capabilities

Agent Skills let you package reusable instructions, scripts, and resources into a directory that Claude loads automatically — across Claude.ai, Claude Code, and the API.

What can be built: PR description generator (reads git diff, outputs structured PR body) + 5 more

MCP

MCP: The USB-C for AI Tools

The Model Context Protocol is an open standard for connecting AI applications to external data and tools — and it has become the de-facto industry protocol since its November 2024 launch.

What can be built: MCP servers exposing internal APIs, databases, or file systems to Claude agents + 5 more

Workflows

Claude in Creative + Technical Workflows

Anthropic has published guidance on using Claude for creative work. Here's what that means for developers building tools at the intersection of design, writing, and code.

What can be built: End-to-end landing page generation (copy + code + structure) from a single brief + 4 more

Notes are published as meaningful Anthropic and Claude updates happen.
Content is based on official sources — no speculation, no invented features.