Shape agent capabilities into your workflow.

skild is a package manager for Agent Skills that installs from Git URLs, local directories, or its registry. It can list, update, validate, remove, and synchronize skills across platforms including Claude, Codex, Cursor, Windsurf, and others, at either global or project scope. Users can also install bundled skillsets, initialize new skills, push them to Git repositories, and publish them through the registry.

GenerateBlocks Skills gives AI coding assistants the block schemas, attribute ordering, dynamic-tag syntax, and layout patterns needed to build WordPress pages with GenerateBlocks. It covers static sections, query loops, ACF content, animations, full-site templates, and workflows for converting HTML, Elementor, or Figma designs. The installer supports several agents, and the repository includes reusable examples. The guidance was checked against Free 2.3 and Pro 2.6 source; some features were still beta, and Pro blocks require the corresponding plugin version.

This repository publishes several OpenCode skills that can be installed individually for different parts of a coding workflow. The current set explores multiple interface directions before implementation, uses the Grok CLI for fresh research, queries local OpenCode history in read-only mode, and helps draft X posts from documented recommendation-system behavior. It is a collection rather than one integrated application, so dependencies, data access, and operating assumptions differ by skill.

Point a coding agent at this repository to identify prompt-cache fixes that match the harnesses you actually use, apply the cited diffs, and run each skill's verification steps. The bundle covers targeted issues in tools such as Aider, Cline, Roo Code, Continue, and OpenCode, backed by separate audit notes. It is not a universal patch set: some audited tools, including Codex CLI and the tested Claude Desktop Code baseline, are documented as already caching correctly.

Honeycomb gives coding agents such as Claude Code, Cursor, and Codex a shared memory that persists across sessions, tools, devices, and teammates. A long-running local daemon receives captures from hooks, CLI, MCP, and SDK clients, then exposes three levels of recall: compact keys, summaries, and full raw sessions. Storage is backed by Deeplake, so setup includes Deeplake authentication; the daemon is the sole storage client and centralizes those credentials behind loopback HTTP.

Senpi Skills is the open-source skills and strategy layer of the Senpi Hyperliquid AI Harness. It packages agent workflows for market analysis, portfolio checks, and signal generation alongside more than 80 strategy templates. Trade execution, exit rules, and risk limits are handled by the separately installed `@senpi-ai/runtime`; self-hosting also requires an OpenClaw host, a Senpi MCP access token, and a funded wallet for each strategy instance.

This collection helps AI assistants handle common Dart and Flutter work through focused guides for architecture, persistence, adaptive interfaces, and testing. Each module explains when to use it and provides an actionable workflow for implementation, refactoring, or troubleshooting, with additional skills for AGENTS.md generation and backend-driven UI.

This Pydantic AI reference project demonstrates a vendor-independent skill system based on progressive disclosure. The agent keeps only brief skill metadata in its initial context, loads a full `SKILL.md` after selecting a relevant skill, and retrieves deeper references or scripts only when required. Included examples cover tasks such as weather, academic search, and code review, with a CLI for trying the runtime. It is primarily a foundation for developers building their own agents and requires local dependencies plus credentials for the chosen model provider.

Computational Chemistry Agent Skills equips OpenClaw with workflows for molecular conversion, conformer generation, DFT and molecular dynamics setup, machine-learning potentials, phonon analysis, reaction networks, and cluster jobs. Users can install the collection or select modules that match their software and research stage. The skills orchestrate existing scientific tools rather than replace them, so each workflow still depends on the required packages, models, licenses, hardware, or HPC access; some modules also need network access for dependencies.

Shokunin bundles 62 engineering skills with ChromaDB-backed persistent memory, MCP servers, subagent configuration, autosave scripts, and a local ecosystem dashboard. Its core skills and memory tools are designed for runtimes including OpenCode, Claude Code, Cursor, Cline, Continue, and Windsurf, with PowerShell and Linux setup paths. Each runtime still needs its own MCP configuration, rule files, and local dependencies applied from the provided templates.

master-skill researches a chosen niche across six tracks: leading practitioners, tool maps, workflows, canonical knowledge, information sources, and terminology standards. It then distills the material into a runnable Agent Skill with companion shell commands, source checks, claim tracing, quality gates, and installation tooling. Generated skills can be installed for Claude Code, OpenClaw, Codex, or Hermes. The repository describes a full research and generation run as taking roughly 30 to 60 minutes.

Codex-Skills offers a curated public core of 70 skills for website and app work, spanning design direction, frontend implementation, motion, accessibility, copy polish, QA, and maintainer safety. You can install the full collection or choose smaller groups for a focused workflow. Codex is the primary target, with documented adapter paths for Claude Code, Cursor, and Gemini CLI, plus local demos and repository-scoped benchmarks for evaluation.

Codex ChatGPT Bridge lets ChatGPT handle scoped planning and review while Codex remains responsible for local execution and verification. It offers OAuth approval, narrow file roots, explicit on/off controls, and token rotation for a Windows-based setup. The trust boundary is significant: an authorized app can read and write files and run shell commands as the local user, so the bridge should stay off when idle and use the smallest possible workspace root.

Webflow Skills gives compatible agents guided workflows for managing CMS content, auditing links and accessibility, optimizing assets, working in Designer, and publishing changes through a preview-and-confirm process. The same plugin also covers Webflow CLI tasks and Code Component creation, deployment, and troubleshooting. Site operations require a Webflow account and configured Webflow MCP server, while CLI and component workflows additionally require Node.js and the Webflow CLI.

WordPress Agent Skills contains early prototypes for creating WordPress themes and sites through Claude Code or Claude Cowork. A user describes the site, reviews specifications and design directions, and the workflow generates a block theme for deployment into a local WordPress Studio site. The accompanying Studio MCP server can create sites, write files, run WP-CLI commands, and produce shareable previews. The project is explicitly beta and experimental; its generated themes are not production-ready and may contain bugs or security issues.

This collection brings familiar thinking and decision frameworks into AI coding assistants, including OODA, pre-mortems, red teaming, Six Thinking Hats, the Feynman technique, Five Whys, RICE, retrospectives, and writing guidance. Users can install the full marketplace or only selected categories or skills in Claude Code, then invoke them with dedicated commands after restarting. It is useful when a task benefits from structured analysis, brainstorming, root-cause exploration, prioritization, reflection, or writing review.

swiftui-skills extracts Apple-authored SwiftUI and platform guidance from the user's local Xcode installation, then makes it available to AI coding agents for generation, refactoring, and review. This helps Claude, Codex, Cursor, OpenClaw, and similar tools follow idiomatic Apple patterns without redistributing the documentation. Installers and native manifests can place the shared skill in several agent environments. Because the documentation is populated during setup, the full workflow requires a local Xcode installation containing the relevant source material.

Seedance2-skill turns a skill-aware agent into a video concept workspace for Seedance. It can work from text, images, video, or audio, branch into several ideas when given only an image, and evaluate prompts for memorability, surprise, emotional movement, and narrative change. A reference library supplies more than one hundred cinematography terms and style cues. Once a concept is ready, the included Python CLI can submit generation requests across Seedance 2.0, 1.5 Pro, and 1.0 model families.