agent-workshop

Skills for designing and evaluating multi-agent systems: orchestrator/worker decomposition, output quality review, and self-improving evaluator/optimizer loops.

Anthropic's official creative and design skills for generative art, frontend design, visual themes, Slack GIFs, and brand-consistent styling.

Anthropic's official development skills for Claude API integration, MCP server building, skill creation, web artifact building, and browser-based testing.

Anthropic's official document processing skills for creating, reading, and editing Word (.docx), PDF, PowerPoint (.pptx), and Excel (.xlsx) files with professional formatting and validation.

Anthropic's official enterprise communication skills for co-authoring documentation and writing internal communications like status reports, newsletters, and FAQs.

Unified quality orchestrator that runs all project linters, formatters, type checkers, and tests as a single pass/fail gate.

Keeps CLAUDE.md files in sync with the actual codebase by detecting meaningful changes and proposing targeted updates.

Evaluation framework skills for designing scoring rubrics, running structured evaluations on LLM outputs, and comparing candidate outputs to recommend a winner.

Skills for persistent memory across Claude Code sessions: initialize a memory file, recall relevant context for the current task, and persist session learnings.

Observability skills for designing logging and tracing strategies, instrumenting existing code with structured log points, and analyzing trace logs to diagnose production issues.

Structured plan lifecycle management with automatic tracking, gate validation, and archival of implementation plans.

Produces phased implementation plans with explicit gates and validating tests at each checkpoint.

Test development and improvement system that tracks quality patterns and progressively improves the test suite.

Automated workspace hygiene checks that detect build artifacts, stale temp files, orphaned docs, and project-level clutter.