Claude Skills by agentskillexchange

exa-js is the official JavaScript SDK for Exa, a web search API built for AI workflows. It gives agents a concrete way to search the web, retrieve page contents, and generate sourced answers from code instead of stitching raw HTTP calls by hand.

Excalidraw is an open-source virtual whiteboard for creating hand-drawn style diagrams, wireframes, and sketches. It provides an infinite canvas, real-time collaboration, end-to-end encryption, and a React component library for embedding in custom applications.

Use generator and validator loops for infra, CI, and platform work so agents ship operational changes with more checks and less improvisation.

ExifTool by Phil Harvey is a comprehensive Perl-based CLI tool for reading, writing, and editing metadata in over 400 file types. It extracts EXIF, IPTC, XMP, GPS, and maker note data from images, videos, audio, PDFs, and documents, making it the industry standard for metadata forensics and batch processing.

Uses DeckTape to open an existing HTML slide deck and render it into a reviewable PDF or per-slide screenshots. It fits agent workflows that need a stable handout, attachment, or archive artifact from Reveal.js, Remark, WebSlides, and similar browser-based presentations.

Convert Obsidian vault content, wikilinks, and attachments into portable Markdown trees before publishing, indexing, or migrating notes.

Use MCP Toolbox to turn database access into a constrained MCP tool surface, with prebuilt or custom tools that agents can call without hand-rolled wrappers for every project.

Mount FastAPI-MCP on an existing FastAPI app so agents can call authenticated route-backed MCP tools with preserved schemas.

Use a FHIR MCP server to query healthcare data resources for documentation and integration workflows while keeping PHI and clinical-use limits explicit.

Turn an existing Haystack pipeline into an HTTP or MCP endpoint without building and maintaining a custom wrapper service.

Use Defuddle when an agent needs clean, metadata-rich article text or Markdown from noisy web pages before summarizing, indexing, or archiving them.

Use bb-browser when an agent needs to inspect pages, pull data, or complete form-driven tasks inside the user's actual logged-in Chrome session. It is for browser work where normal HTTP fetches fail because the important context lives behind auth, cookies, or internal web apps.

Uses invoice2data to turn invoice PDFs into structured JSON, CSV, or XML using supplier-specific templates. This is for repeatable invoice field extraction and renaming workflows, not for full accounting system automation or generic OCR catalog listings.

Watch Claude Code, Codex CLI, Cursor, or OpenCode sessions, extract durable project memory, and keep it locally as reusable markdown.

Uses extruct to pull machine-readable metadata from raw HTML so an agent can classify, deduplicate, or enrich pages without brittle full-page parsing. It is best for metadata harvesting workflows, not for crawling an entire site or rendering JavaScript-heavy pages.

Open a rendered page, describe the fields you want, and return structured data without hand-maintaining brittle selectors.

Use MailParser when an agent receives raw RFC822 or MIME email and needs a normalized result with headers, text, HTML, addresses, and attachments. This is for email ingestion and handoff workflows, not for acting as a mailbox client or delivery platform.

Use Xidel to pull targeted values from pages, XML documents, or JSON APIs with CSS selectors, XPath, XQuery, or JSONiq expressions.

Convert PDFs into LLM-ready markdown or coordinate-aware JSON, and use the same pipeline for tagged-PDF accessibility workflows when that is the real job to be done.

Expose one document-extraction surface to MCP-compatible agents so they can normalize PDFs, Office files, images, HTML, and other mixed inputs before downstream review or indexing.

Use i18next-parser when an agent needs to scan a codebase, find translation calls, and update locale resource files as part of localization maintenance. This is a bounded catalog-maintenance workflow, not a generic i18n platform listing.

A modern, maintained replacement for the ls command, written in Rust. eza provides colorized output, Git integration, tree views, symlink awareness, and extended attribute support in a single fast binary.

Fabric is an open-source framework for augmenting humans using AI. It provides a modular system of crowdsourced prompt patterns that solve specific problems—from summarizing content to extracting wisdom to analyzing security threats—all usable from the command line.

Monitors container runtime events using Falco sysdig libraries and sends alerts on suspicious syscall patterns. Integrates with Kubernetes audit logs and Prometheus AlertManager for real-time threat detection.

Use the upstream install or setup path that matches your environment: - cmake \ - make -j$(($nproc-1)) falco_unit_tests;

faster-whisper is SYSTRAN’s high-performance reimplementation of OpenAI Whisper on top of CTranslate2. It is built for transcription pipelines that need lower latency, lower memory usage, optional quantization, and practical Python integration for batch or real-time speech workflows.

FastMCP is the standard Python framework for building Model Context Protocol servers, clients, and applications. It provides automatic schema generation, transport negotiation, and protocol lifecycle management, letting developers wrap Python functions into MCP-compliant tools with minimal boilerplate.

Faust.js by WP Engine is a JavaScript framework and WordPress plugin that streamlines building headless WordPress sites with Next.js. It handles data fetching via WPGraphQL, authentication, content previews, and server-side rendering, providing a production-ready bridge between WordPress as a CMS and React-based frontends.

A fast and user-friendly alternative to the find command, written in Rust. fd provides intuitive syntax, regex and glob support, colorized output, parallel execution, and automatic .gitignore awareness.

Feroxbuster is a high-performance content discovery tool written in Rust that performs forced browsing attacks to enumerate hidden files, directories, and endpoints on web servers. It features recursive scanning, smart filtering, and extensive output options.

Normalizes audio loudness to broadcast standards using FFmpeg loudnorm filter with EBU R128 two-pass analysis. Measures integrated LUFS, true peak, and loudness range via ffmpeg -af loudnorm=print_format=json.

Transcodes and processes audio files using the FFmpeg CLI and libavcodec library. Supports batch format conversion, loudness normalization via EBU R128, and metadata extraction with ffprobe.

Orchestrates parallel FFmpeg transcoding jobs with hardware-accelerated encoding via NVENC, VAAPI, or VideoToolbox. Supports HLS adaptive bitrate packaging using ffmpeg -f hls with multiple -map streams and fmp4 segment formatting for DASH output.

Batch transcode media files using FFmpeg CLI with preset profiles for web, mobile, and broadcast delivery. Supports hardware-accelerated encoding via NVENC/VAAPI and automated quality analysis with VMAF scoring.

Extracts video clips and segments using FFmpeg libavformat and libavcodec APIs. Supports keyframe-accurate cutting with -ss/-to flags, stream copy mode, and re-encoding via libx264/libx265 presets.

Orchestrates complex video and audio processing pipelines using FFmpeg and FFprobe with scene detection via PySceneDetect. Handles format conversion, thumbnail generation, HLS packaging, and loudness normalization per EBU R128.

Automated video and audio transcoding using FFmpeg with hardware-accelerated encoding via NVENC/VAAPI, HLS adaptive streaming output, and MediaInfo-based quality validation.

Generates contact-sheet-style thumbnail grids from video files using FFmpeg tile filter and libvips. Supports customizable grid dimensions, timestamp overlays, and batch processing across directories.

Creates video thumbnail mosaics and sprite sheets using FFmpeg filters and the fluent-ffmpeg Node.js wrapper. Generates contact sheets, animated GIF previews, and WebVTT thumbnail tracks.

Builds complex FFmpeg filtergraph chains for batch video transcoding, thumbnail sprite generation, and HLS adaptive bitrate packaging. Supports NVIDIA NVENC hardware acceleration and HDR tone mapping.

ffsubsync automatically aligns subtitle files to spoken audio by analyzing timing from a reference video or audio file. It is a strong fit for agent workflows that need to repair drifting subtitles without manual waveform editing.

Extracts design tokens from Figma files using the Figma REST API v1 and transforms them into platform-specific formats via Style Dictionary. Supports CSS custom properties, Tailwind config, and iOS/Android native tokens.

Extracts design tokens from Figma files using the Figma REST API and style-dictionary. Generates CSS custom properties, Tailwind configs, and Swift/Kotlin theme files from Figma components.

Exports assets from Figma files using the Figma REST API v1. Extracts component sets, design tokens (colors, typography, spacing), and renders frames as SVG/PNG. Generates CSS custom properties from styles.

Exports design assets from Figma files using the GET /v1/files/:key and /v1/images/:key endpoints. Supports SVG, PNG, and PDF export with scale and format parameters.

FilePond is a JavaScript file upload library that handles files, directories, URLs, and Data URIs with drag-and-drop, paste, and API support. It includes built-in image optimization, async chunked uploads, and adapters for React, Vue, Angular, and Svelte.

files-to-prompt by Simon Willison concatenates an entire directory of files into a single prompt for use with LLMs. It supports file extension filtering, gitignore-aware exclusions, Claude XML format output, Markdown fenced code blocks, line numbering, and stdin piping for flexible codebase-to-prompt workflows.

Screen prompts and responses with input and output scanners before an LLM interaction reaches production users or downstream systems.

Use Better Icons when an agent needs to search icon sets, compare matches, and return the exact SVG asset needed for a UI task. It is a narrow asset-selection skill, not a generic icon platform listing.

Scan large directory trees for duplicate or under-replicated files, then review grouped results before cleanup actions.