Automate Chrome tasks such as navigating pages, clicking buttons, and filling forms. Capture screenshots or GIF recordings, and inspect console logs, the messages a website produces, to review interactions and investigate problems.
Import an existing React component library into Claude Design, including its styles, usage references, and previews. This lets the design agent build screens and prototypes using the same components already maintained in the source project.
Review code changes in a diff, pull request, branch, or file path to identify bugs and cleanup opportunities. Findings include file locations and concrete failure scenarios, with options to apply fixes or post comments.
Review pending code changes for newly introduced security vulnerabilities with clear exploitation paths. Focus on concrete, significant risks and filter likely false alarms, then report affected files, line numbers, severity, attack scenarios, and recommended fixes in Markdown.
Convert an HTML slide deck into a PowerPoint file with editable text, shapes, and images. The skill handles font substitutions, imports speaker notes, and checks export warnings to help identify missing images or incorrectly captured slides.
Create, read, and edit Word documents and templates, including image replacement, comments, and tracked changes that record revisions. The workflow also checks document structure and renders finished files for visual review.
Read, create, and organize PDF files, including merging documents, splitting pages, extracting tables, and filling forms. It also covers password protection and OCR, which recognizes text in scanned pages for extraction or search.
Set up Cowork for everyday work through guided steps: choose plugins suited to a role, connect work tools, try a skill, and optionally save a writing voice profile for future drafts.
Create charts and dashboards with a consistent approach to chart selection, color validation, labels, and interaction. Includes a palette checker and accessibility rules for readable visualizations in both light and dark modes.
Create or improve a project-specific skill with verified build, launch, and interaction steps. Package concise instructions with scripts or existing command-line tools so future agents can run and exercise the application.
Build and debug mods that change Claude’s interface or behavior. Add panels, status messages, slash commands, or hooks that respond to prompts and tool calls, with guidance for file structure, type checks, and behavior tests.
Create and improve agent skills by defining tasks, writing instructions, and testing realistic requests. Compare outputs, success rates, time, and token use, then refine the skill and its description so it triggers appropriately.
Coordinate large migrations, refactors, and bulk renames by splitting them into independent tasks. After plan approval, 5–30 agents work in separate Git worktrees, isolated working directories, then review, test, and open a pull request for each task.
Create a 3D object with three.js, inspect it from every angle in an interactive viewer, and download it as OBJ + MTL or GLB for use in other 3D applications.
Build Claude applications using guidance for the official Anthropic SDK. Choose an integration approach, implement streaming and tool use, and follow workflows for model migration, prompt audits, evaluation, and cost optimization.
Check whether a code change works by running the app and using its real interface. Exercise the intended flow and nearby error cases, capture outputs or screenshots, and report observed behavior with a clear verdict.
Is this your account? Link GitHub to bring it onto your creator page (Japanese)