Claude Code ecosystem. Plugins, skills, workflows, hooks, and tools worth understanding.

The official Claude Code repository, release history, changelog, issue tracker, and distribution source.

Best for

Following releases, checking known issues, and tracing a product change back to Anthropic.

Why it is interesting

It is the primary public change stream behind Claude Code and the first stop for version-specific research.

Check before installing

Use the official documentation for supported behaviour; issues and discussions can describe unconfirmed or version-specific experiences.

Anthropic's official directory of Claude Code plugins, including Anthropic-maintained and third-party entries.

Best for

Discovering plugins that use the current Claude Code marketplace format.

Why it is interesting

Its marketplace manifest is a live catalog and a practical reference for plugin structure, metadata, and distribution.

Check before installing

Official directory inclusion is not a guarantee that third-party code is safe. Review each plugin's hooks, MCP servers, scripts, permissions, and source repository.

A skills-based software-development workflow covering design, planning, testing, debugging, review, and task execution.

Best for

Developers who want a strong, repeatable engineering process around an agent rather than ad-hoc prompting.

Why it is interesting

It treats skills as an end-to-end development methodology and supports several coding-agent runtimes from one repository.

Check before installing

It changes how the agent approaches most development work. Review its hooks and skills first; the optional visual companion also documents opt-out telemetry controls.

A skill and plugin that asks coding agents to give shorter, compressed responses while preserving technical details.

Best for

Users who find long agent explanations distracting or want to reduce conversational output tokens.

Why it is interesting

It packages response style, modes, lifecycle hooks, status-line feedback, and reproducible benchmarks as one cross-agent project.

Check before installing

The project distinguishes chat-response savings from full coding-run savings. Review the benchmark scope and ensure compressed prose remains clear for security and destructive actions.

A plugin that pushes coding agents toward the smallest adequate implementation by checking reuse and native options before adding code.

Best for

Teams fighting over-engineering, speculative abstractions, unnecessary dependencies, or oversized diffs.

Why it is interesting

Its decision ladder turns YAGNI and reuse into an always-on agent behaviour, with audit and review commands for existing code.

Check before installing

The Claude Code plugin uses lifecycle hooks and requires Node on the non-interactive PATH. Minimalism should not remove required validation, accessibility, or safety controls.