Top Agent Skills

Provides guidance and best practices for working with Cassandra, covering setup, configuration, development, deployment, and operations.

Routes freeform user requests to the appropriate GSD command based on intent, confirming the routing before dispatching.

Defines brand voice, visual identity, messaging frameworks, and asset management, with scripts to sync guidelines to design tokens and validate assets.

Builds reliable browser automation skills by iteratively running a browsing task, reading the trace, and improving the navigation strategy until it passes consistently.

Guides the creation of high-quality MCP servers, covering design, implementation, testing, and evaluation for Python and TypeScript.

Runs comprehensive SEO audits and analysis for websites, covering technical SEO, content quality, schema markup, Core Web Vitals, local SEO, and AI search optimization using browser automation tools.

Adopts the persona of Diogenes of Sinope to deliver provocative, stripped-down advice and challenge assumptions through radical simplicity and irreverent wit.

Provides a strategic advisory persona based on A.G. Lafley, applying the Strategy Choice Cascade and consumer-centric frameworks to business decisions.

Automates the full build-to-launch cycle for Blender extensions and add-ons, including packaging, installation, enabling, and launching Blender for rapid testing.

Generates a codebase architecture document and CLAUDE.md scaffold by exploring source code and asking targeted questions.

Guides a context-driven release workflow with research, planning, verification, and autonomous execution phases.

Lists Claude Code components such as skills, environment variables, sessions, settings, and tools, with options to research or display configured values.

Work incrementally on issues using Git worktrees for parallel agent workflows, starting each session by getting bearings and ending with clean state.

Analyzes tabular datasets to understand distributions, data quality, and patterns, including column distributions, missing data, class balance, and summary statistics.

Configures Rei Qwen3 Coder as a model provider for Clawdbot, including setup scripts, model switching, and troubleshooting 403 errors.

Configures and operates moonrepo monorepo build systems, covering workspace and project configuration, task orchestration, toolchain management, CI integration, Docker support, and querying.

Manages conversation context across sessions and compactions using TLDR summaries, structured documentation, and searchable history.

Create, update, and comment on GitHub pull requests with a focus on material impact, safety, and human reviewability.

Plans features interactively, creates dependency-ordered tasks, and prepares Ralph for autonomous execution.

Create, validate, and improve ALPS profiles for RESTful API design, generating from natural language and providing validation and improvement suggestions.

Manage Helm charts end-to-end: structure, templates, values schema, releases, hooks, and debugging.

Build AI agents, multi-agent teams, and agentic workflows using the Agno framework, with reference files covering agents, teams, workflows, memory, knowledge, and deployment.

Write, read, and validate HUML documents, converting between YAML/JSON/TOML and HUML for human-readable configuration files.

Coordinates a suite of Wolf agent skills covering principles, archetypes, governance, roles, and verification, with mandatory initialization and sequential skill chaining.

Interact with Atlassian Jira via REST API, JQL, workflow transitions, sprints, and webhooks for issue tracking and project management.

Optimizes Flutter app performance by reducing rebuilds, improving list rendering, managing memory, and applying profiling and image optimization techniques.

Implements feature flags and remote configuration in Flutter apps using Firebase Remote Config and LaunchDarkly, enabling staged rollouts, A/B testing, and kill switches without app updates.

Implements in-app purchases and subscriptions in Flutter using the in_app_purchase plugin and RevenueCat, covering consumables, non-consumables, restore purchases, and receipt validation.

Integrates WebViews in Flutter apps using flutter_inappwebview, covering HTML content loading, JavaScript bridges for bidirectional communication, navigation handling, cookies, and storage.

Automatically answers Gmail messages on behalf of a client.

Participate in anonymous imageboard discussions with other AI agents via a JSON API, including browsing boards, threads, and posting replies.

Tracks media progress, reviews, collections, analytics, and calendar in Ryot, with automation support.

Guides users through creating, testing, and deploying browser automation functions with the Stagehand CLI, including interactive site exploration, coding, and deployment to Browserbase.

Produces a machine-validated, auditable JSON decision record with assumptions, risks, recommendation, and review gating for high-stakes decisions.

Plans and decomposes complex, multi-step research tasks into manageable steps.

Wraps Claude relay sessions with a short slash command for quick invocation.

Answers questions about the Sui blockchain ecosystem, covering concepts, tokenomics, validators, staking, and comparisons with other chains.

Design and implement Oban background job workers for Elixir, covering queue configuration, retries, uniqueness, cron scheduling, and testing.

Controls the Winamp media player on Windows, WSL2, or Linux via Wine using command-line switches for playback, playlists, and instance management.

Query and monitor UniFi network devices, clients, health, top applications, and alerts via the local UniFi OS gateway API.

Manage Remember The Milk tasks — list, add, complete, delete, search, prioritize, tag, move, and annotate tasks.

Surveys a codebase as a senior advisor and produces prioritized, self-contained implementation plans for other models or agents to execute, without modifying source code.

Tracks Jira issues, projects, and sprints via the Atlassian API, including discovery, JQL search, transitions, and worklog management.

Sends email messages via Gmail with support for attachments, CC/BCC, HTML content, send-as aliases, drafts, and dry-run previews.

Generates new design screens from text or images, edits existing screens with prompts and design tokens, and creates design variants using Stitch MCP, with prompt enhancement and asset downloading.

Creates a new Google Tasks list and populates it with initial tasks using the gws CLI.

Geocode addresses, find nearby places, compute routes and distances, and get timezones using free OpenStreetMap data with no API keys.

Capture and manage Today I Learned entries on OpenTIL directly from the CLI, including drafting, publishing, editing, searching, and syncing entries.

Tracks async and multi-step work as JSON tasks in .dex/tasks/, with commands to create, list, show, complete, edit, and delete tasks.

Searches, reserves, and manages Korean SRT train tickets via CLI, including continuous monitoring for canceled seats.

Orchestrates VPN switching, geo-block diagnostics, and leak-aware task resumption by coordinating shell, curl, WireGuard, Tailscale, DNS, IPinfo, and MoltGuard skills.

Query Google Search Console for SEO data including search queries, top pages, CTR opportunities, URL inspection, and sitemaps to analyze search performance and indexing status.

Blocks dangerous shell commands and warns on risky ones, applying a safety protocol before executing any command.

Queries Korean public holidays and performs business day calculations via the HuDy MCP server, including counting business days and managing custom holidays.

通过 paramiko 库建立 SSH 连接，在远程服务器上执行命令、安装软件、查看日志，并支持交互式会话。

Search, analyze, and monitor X/Twitter from the terminal with a fast Rust CLI, including AI-powered analysis, trends, and follower tracking.

Mint a unique 1/1 villain NFT from CHUM's agent-only collection on Solana, with a 0.001 SOL mint fee plus network fees.

Writes documents in a plain, human voice and removes phrasing that reads as machine-generated, applying a two-pass write-and-audit process to any written artifact meant to be read by a person.

Generates interactive HTML explainers for code diffs, branches, and PRs from a single Markdown document, using a renderer to produce styled output with diagrams and quizzes.

Builds a navigable knowledge graph from any folder of files, with community detection, an audit trail, and outputs including interactive HTML, GraphRAG-ready JSON, and a plain-language report.