Coding standards, frameworks, refactoring, and code review (16.4K skills)
Build MCP servers, which give AI agents tools for external services, using Python or TypeScript. The skill guides tool design, input validation, error handling, code review, and evaluations based on realistic tasks.
Fetch execution traces, records of an agent’s actions, from LangSmith Studio to debug LangChain and LangGraph agents. Inspect errors, tool calls, memory operations, execution times, and token usage, or export records for further review.
Create or improve Claude skills that package specialized knowledge, repeatable workflows, and tool instructions. The guide covers planning SKILL.md and supporting files, generating a starter folder, validating the structure, and packaging the result as a distributable ZIP.
Add support for a programming language to codegraph, from connecting its parser and implementing code extraction to writing tests. Evaluate extraction quality and compare agent effort with and without codegraph across three real codebases.
Review slow page loads and choose resource hints, browser instructions that start loading or connecting earlier. Use measurements to identify bottlenecks, then recommend preload, preconnect, prefetch, or dns-prefetch for important assets and likely next-page resources.
Review HTML object elements and add meaningful fallback content for people who cannot load or view embedded material. Identify missing alternatives, suggest text or appropriate HTML elements, and explain how to verify accessibility.
Review website images for high-density displays, where standard assets can look blurry. Check higher-resolution versions, HTML and CSS image selection, delivery settings, and loading behavior, then identify specific files or components that need changes.
Review early connections to third-party services such as fonts and APIs. Use loading measurements to identify essential domains, recommend preconnect hints that prepare connections in advance, and remove unnecessary hints that add overhead.
Review stylesheets for a CSS reset or normalization layer that reduces differences in browser defaults. Check import order, framework provisions, and rendered layouts across screen sizes and interaction states, then suggest changes for a consistent styling foundation.
Review pages for situations where JavaScript is disabled or unavailable. Check that important content appears in the initial HTML and that noscript elements offer useful alternatives, then suggest changes to keep core content and navigation accessible.
Review picture elements that select images by format or display conditions. Check fallback images, alternative text, dimensions, and source ordering, then provide corrected HTML with JPEG or PNG fallbacks for browsers that cannot use the selected formats.
Review images displayed at different screen sizes and help browsers select appropriately sized files. Check srcset image options, sizes display-width hints, and CSS layout together to identify oversized mobile downloads or mismatched image variants.
Measure the total size of a web page and its resources, identify what slows loading, and reduce unnecessary weight. Use a 1500KB limit to guide image optimization, code compression, and removal of unused resources.
Check whether a web app is ready for PWA installation, which lets users install it like an app. Review HTTPS, app settings, icons, and offline responses, and identify missing or inconsistent configuration.
Find CSS and JavaScript that delay a page’s first visible content. Confirm the bottleneck with browser tools or performance data, then adjust loading behavior and inline essential styles to help the page start rendering sooner.
Create a custom offline page and configure a service worker, a browser background script, to cache it in advance. Show it when page navigation fails, and review resource dependencies and fallback conditions.
Review website JPEGs for progressive encoding, which displays a blurry preview before sharpening. Identify images that could use this format, consider file size and loading behavior together, and explain how to confirm the change in browser developer tools.
Audit web images that are larger than their displayed size and help serve suitable versions for different screens. Review image dimensions, responsive image settings, and Lighthouse findings, then provide updated HTML and steps to verify the changes.
Review pagination markup so people using screen readers or keyboards can understand their current page and move between pages. Check navigation labels, current-page attributes, focusable controls, and announcements when the page changes.
Review messages that display counts and check whether each target language gets the correct plural form. Replace simple number-and-text joins or singular/plural branches with Intl.PluralRules or ICU MessageFormat, which select wording according to language-specific rules.
Audit page loading and choose resource hints, browser instructions for fetching files or preparing connections early. Use measurements to identify delays, then suggest preload, preconnect, prefetch, or dns-prefetch for critical assets, external services, and likely next-page resources.
Review image loading settings based on where images appear on a page. Identify offscreen images that should load later, keep important visible images loading promptly, and add image dimensions to reduce layout shifts.
Review fixed pixel values in fonts, widths, and spacing, then suggest relative units that adapt to available space and browser font preferences. Check rendered layouts at different screen widths and during interactions before proposing changes.
Measure page loading speed and identify what prevents a page from loading in under three seconds. Use performance measurements to guide improvements such as lazy loading, caching, CDN delivery, and reducing resource costs.