Shape agent capabilities into your workflow.

These Datadog skills let compatible agents manage monitors, search logs and traces, analyze agent-observability experiments, investigate CI failures and flaky tests, and work through audit-trail incidents. Individual workflows use dd-pup, Datadog MCP toolsets, or REST APIs, so the required account access, permissions, and API credentials vary by skill.

Kapso Agent Skills covers three WhatsApp workflows: integrating accounts, webhooks, messages, templates, and flows; building automations with agents, functions, and databases; and inspecting logs or webhook deliveries to diagnose failures. The skills prefer an authenticated Kapso CLI when available, with scripts and direct API calls as fallbacks. The collection is still in alpha and requires Kapso API configuration and credentials.

This Claude Skill turns Markdown files, chat summaries, and research reports into downloadable EPUB3 ebooks. It detects chapters from headings, builds a navigable table of contents, preserves common Markdown formatting, and styles code blocks, tables, and metadata for reading devices. The output works with standard EPUB readers including Apple Books, Kindle apps, Google Play Books, and Kobo; running the included scripts locally requires the listed Python dependencies.

Use focused skills for TileLang GPU kernel development, Megatron memory estimation, SLIME reinforcement-learning training, and supporting tasks such as technical diagrams or presentations. Some skills are complete while planned SGLang and vLLM modules are not yet available. The repository is under active development and much of its content was generated with LLM assistance, so code and recommendations should be verified before production use.

Repo Forensics scans untrusted agent repositories locally for prompt injection, malicious packages, MCP tool poisoning, credential access, data exfiltration, obfuscated execution, and compound attack chains. It can audit an existing agent stack or install hooks that block known package indicators before commands and run a broader review after clones or installs. Detection rules and vulnerability intelligence can be updated separately, and the scanner reports no telemetry.

dash_skills is a personal collection of Agent Skills for Dart and Flutter coding assistants. It covers areas such as language best practices, documentation validation, testing and coverage, package maintenance, long-line handling, and command-line profiling. Skills can be installed individually or together through the Claude Code plugin marketplace, but this repository is separate from the official Dart and Flutter skill collections.

Skills-4-SE organizes more than 180 Claude skills around practical software engineering work, including code generation, testing, quality analysis, documentation, architecture, DevOps, collaboration, and maintenance. Developers can browse the catalog and add narrowly scoped workflows to an assistant instead of relying on one broad prompt for every engineering task.

Claude Agent Team Manager turns existing Claude Code agents and skills into a visual, drag-and-drop organization chart. Users can save team configurations, share variables, generate a deployment primer from a short objective, and launch the prepared run in a terminal. Teams and chained pipelines can also be scheduled through the operating system; the macOS build is unsigned and needs manual approval on first launch.

Marketing Skills combines more than 51 portable agent workflows with a TypeScript command-line platform for marketing work. The Markdown skills cover conversion optimization, copywriting, SEO, analytics, advertising, and growth tasks across several agent hosts, while the unified CLI exposes JSON-oriented access to dozens of external marketing integrations. Redis-backed response caching is optional, and each connected service still requires its own user-provided credentials.

This starter pack combines four standalone 2D and 3D game projects, a bonus isometric-sprite workflow, eight game-development agent skills, and beginner tutorials. Builders can begin with Phaser platformer or tactics examples, Three.js arena and mini-game templates, or the asset workflow instead of starting from an empty folder. The repository root must be opened for the shared skills to be visible to the agent, and some starters—such as the Oak Woods project—require external art assets.

OpenClaw Marketing Skills packages 37 workflows across conversion optimization, copywriting, SEO, paid ads, email, growth, retention, sales, and go-to-market work. A shared product-marketing context keeps later tasks grounded in the same product and audience instead of repeating setup. Optional connectors can bring in live Google Ads, Search Console, Meta Ads, and X signals, but those workflows require the corresponding accounts, APIs, or external services.

Awesome Ascend Skills organizes Huawei Ascend NPU development knowledge into installable agent skills for environment checks, model migration and conversion, inference deployment, distributed training, profiling, and operator work. Users can start with a base bundle, add a domain bundle, or copy only the individual skills needed for a project. It is a workflow and knowledge collection rather than a runtime, so real tasks still depend on the relevant Ascend hardware, software stack, and external tools.

This repository packages Agent Skills for developers working with Payload CMS. One skill provides focused guidance for collections, fields, hooks, access control, queries, database adapters, and plugin development, while another leads an interactive migration workflow from systems such as WordPress, Contentful, Strapi, or other exported CMS data. The skills can be installed with the standard `npx skills add` flow and used inside a compatible coding agent. They are instructional workflows for building and migrating Payload projects, not a standalone CMS application.

humanizer-ru is an agent skill for revising Russian prose that sounds bureaucratic, translated, or recognizably formulaic. It checks 54 writing patterns, applies 20 hard bans, runs a four-pass audit, and can calibrate the result to an author's voice while comparing before-and-after metrics. The same SKILL.md format is packaged for Claude Code, Codex CLI, Cursor, Gemini CLI, and other compatible agents. It is aimed at Russian copywriting, UX writing, and general editorial cleanup.

amux is a terminal UI for running multiple coding agents in parallel. Each agent gets its own tmux session, and workspaces can use repository checkouts or Git worktrees so branch changes stay isolated. The same interface can launch agents, show diffs, and open a terminal; it supports Linux and macOS and requires tmux 3.2 or newer.

ClawFu packages positioning, copywriting, persuasion, sales, SEO, customer research, revenue operations, and other marketing methodologies as agent-readable skills. Users can connect the npm MCP server to Claude, Copilot, Cursor, or Windsurf, or consume individual SKILL.md files through compatible tools. The collection is intended for teams that want agents to follow named methods, clear triggers, and defined output formats instead of relying on one-off prompts.

lark-todo is an agent skill that scans messages, meetings, calendars, document comments, approvals, tasks, and mail across one or more Lark organizations, then deduplicates and ranks the pending work by urgency. After the user selects an item, it can draft replies, process approvals, answer mail, respond to invitations, or create tasks, with every write action shown for confirmation first. Initial setup requires Node.js, a Lark custom app, and authorization through lark-cli.

Open Agent Skills is a lightweight, provider-agnostic framework for combining domain instructions with MCP tools across GitHub, filesystem, and browser tasks. It extends MCPMark with skill-based runs so developers can compare task completion and context usage under the same evaluation setup. Running it requires Python 3.11+, model credentials, and any task-specific services or containers. The project describes itself as a starting point, and some included skills remain narrowly tailored, so scripts should be reviewed before production use.