Extract text, forms, and structured data from PDF documents. Supports OCR for scanned documents and complex layouts.

Create, edit, and format Microsoft Word documents. Supports templates, styles, tables, and image embedding.

Work with Excel spreadsheets. Create charts, formulas, pivot tables, and complex data analysis.

Design and generate professional PowerPoint presentations with custom themes, animations, and layouts.

Automated testing for web applications using browser automation. Supports E2E tests, visual regression, and accessibility checks.

Enforces RED-GREEN-REFACTOR cycle: write failing test, watch it fail, write minimal code, watch it pass, commit.

Socratic design refinement through natural dialogue. Explores alternatives, presents designs in sections for validation.

4-phase root cause process including root-cause-tracing, defense-in-depth, and condition-based-waiting techniques.

Fast iteration with two-stage review: spec compliance check, then code quality review. Enables parallel agent workflows.

Creates detailed implementation plans with bite-sized tasks (2-5 minutes each). Every task has exact file paths and verification steps.

Build Model Context Protocol (MCP) servers quickly. Generates boilerplate, handles tool registration, and testing.

Meta-skill for creating new skills. Follows best practices, generates SKILL.md templates, and sets up testing.

Generate beautiful, consistent UI themes. Supports dark mode, accessibility, and multiple color palettes.

Create ATS-optimized resumes tailored to specific job descriptions. Supports multiple formats and styling.

AI-powered PPT image and video generation with intelligent transitions and interactive playback.

Guide for creating autonomous subagents, defining system prompts and tools for Claude Code plugins.

Integrating Model Context Protocol (MCP) servers (stdio, SSE) into Claude Code plugins.

Step-by-step guide for authoring new Agent Skills for Claude Code following .SKILL.md standard.

Implementing Pre/PostToolUse hooks to validate or automate agent actions in Claude Code.

Creating custom slash commands with YAML frontmatter and interactivity for Claude Code.

Writing standardized docstrings following PyTorch engineering conventions.

Extract financial metrics like Net Sales from complex PDF 10-K reports. A benchmark for document processing.

Finalize mathematical proofs using the Lean 4 theorem prover. Tests formal reasoning capabilities.

Analyze proteomics data from Excel sheets for cancer cell line research. Bioinformatics benchmark.

Generate community pulse write-ups and analytics for GitHub repositories using gh-cli.

Perform OCR on images and run statistical analysis on extracted data. Multi-modal benchmark.

Automate the generation of HR offer letters using Word/Docx templates. Document automation benchmark.

Identify discrepancies between PDF archived records and Excel databases. Data comparison benchmark.

Simulate open Dicke models and steady-state quantum systems. Quantum physics benchmark.

Solve for economically efficient power network dispatch. Energy optimization benchmark.

Construct constrained travel itineraries with budget, time, and preference constraints.

Act as an embodied agent to plan household tasks in a virtual environment. Robotics benchmark.

Solve Flexible Job Shop Scheduling optimization problems. Manufacturing benchmark.

Detect exoplanet periods from TESS light curve astronomical data. Astronomy benchmark.

Configure continuous fuzzing environments for Python projects. Security benchmark.