Shape agent capabilities into your workflow.

Light Skills is an agent-skill pack for research, competitions, and innovation projects. It can take a vague direction through literature search, idea generation and critique, research planning, data and experiments, result analysis, scientific figures, paper writing, venue preparation, and intellectual-property materials. A project ledger records goals, decisions, outputs, unverified claims, and next steps so work can continue across conversations. The workflow is deliberately staged, with user checkpoints and rollback paths instead of asking an agent to run the entire project unattended.

Lazyweb Skill Pack gives coding agents evidence-based design research before they build or critique an interface. Its local skills and hosted MCP can search real mobile and desktop screenshots, compare visual references, retrieve multi-screen journeys, guide screen design, map product flows, and propose reviewable changes. Installation and workflow discovery are broadly available, while data-bearing tools and limits depend on the account plan and experiment assignment. An optional autorouter edits only a marked instruction block after consent.

GodotPrompter gives coding agents on Godot 4.3+ game projects on-demand domain knowledge for setting up projects, implementing gameplay, and shipping games, with Godot best-practice guidance and GDScript/C# examples. It works across several agent hosts, while skills for third-party addons require those addons to be installed in the target project.

Designer Skills encodes an eight-step product design workflow for AI coding tools. It can interrogate an incomplete idea, produce a design brief, define information architecture, generate light and dark tokens, split work into buildable slices, implement a mobile-first frontend under a named aesthetic philosophy, and review code or screenshots against the brief. Each stage inspects existing components, themes, and dependencies first. Documents persist by feature in a project-local .design folder, allowing the workflow to resume later.

Education Agent Skills Library contains 165 pedagogical skills across 20 domains. The first 19 support teachers and learning designers, while the twentieth defines live interaction patterns for how an AI should respond to learners during study sessions. It works with Claude, Codex, Hermes, and compatible tools; local or plugin installation is the recommended free path, while the hosted MCP endpoint requires an access token.

VexJoy Agent provides a single `/do` or `$do` entry point that selects a domain agent and a matching workflow skill from a plain-language request. The routed task proceeds through explicit phases with test, review, and verification gates backed by hooks and deterministic scripts. The toolkit covers software work as well as research, writing, data analysis, moderation, and self-contained HTML artifacts.

SkillDeck provides a native three-pane interface for managing skills across Claude Code, Codex, Gemini CLI, Cursor, and other agents. It can browse skills.sh and ClawHub, import from GitHub or local folders, check for updates, edit SKILL.md files, and assign installations through symlinks. It requires macOS 14 or newer, and downloaded unsigned builds may need manual approval on first launch.

Analyze a repository to create or improve project-specific CLAUDE.md files, then keep them aligned as the codebase changes. ClaudeForge can prune stale references, repair linked instruction files, enforce a 150-line limit through modular files, run link and dependency audits, and convert the instruction tree into AGENTS.md; its full hook workflow requires recent Claude Code versions.

Install and update Agent Skills from GitHub, local folders, or ZIP files, validate them against the specification, and manage their metadata from one toolkit. SkillPort can deliver skills through a CLI or MCP server and uses search-first, on-demand loading for large collections. It is still under active development, so its APIs may change.

ConvexSkills provides installable guidance and templates that help coding agents implement common Convex backend work, including functions, realtime data, scheduled jobs, webhooks, and migrations. It supports standard `.agents/skills` and project-local skill directories; the repository also directs users to a separate official Convex Agent Plugins collection.

Claude Code ASO Skill turns App Store Optimization for iOS and Android into guided agent workflows. It can prepare keyword research, metadata drafts, action lists, reports, and executive summaries. Developers can install the coordinated multi-agent system in Claude Code, while other users can upload the standalone ZIP skill to Claude Desktop or the web app for conversational analysis. Some workflows fetch live store data and therefore require an internet connection; app teams should still review the resulting positioning and store copy before use.

Nexscope provides 53 Amazon Seller Skills for OpenClaw, Claude Code, Cursor, Codex, and other compatible agents. They cover keyword research, listing optimization, FBA calculations, PPC campaigns, competitor analysis, sales estimation, and related seller workflows. The skills are Markdown instructions that require no API key, but they cannot access live Amazon marketplace data on their own.

Palkan's skills help coding agents examine Rails applications through the layered architecture principles described in Layered Design for Ruby on Rails Applications. The workflows can audit service objects, model callbacks, god objects, layer boundaries, and code changes, then support incremental refactoring plans and specification tests. Claude Code users get packaged commands and sub-agents through the plugin, while the underlying SKILL.md content and references can also be used from environments such as Codex.

Superpowers Lab hosts experimental Claude Code skills for tasks that are awkward to handle manually. Current examples include finding semantically duplicated functions, invoking MCP tools on demand without permanent configuration, and controlling interactive programs through tmux; the skills are still being refined and may change.

paper2patent provides prompt templates and an agent skill for transforming a research paper into a Chinese invention-patent application draft. It guides the model through the abstract, up to ten claims, the specification, and black-and-white flowcharts or block diagrams under explicit formatting rules. Supporting scripts can generate figures, export DOCX or PDF files, and check whether a draft introduces content not supported by the source paper. The workflow can be loaded into Codex, Claude Code, Cursor, Windsurf, and similar skill-aware tools.

Install awesome-novel in Claude Code, start it from a novel folder, and let a coordinator guide worldbuilding, character setup, volume and chapter planning, drafting, cleanup, review, and archiving. The project keeps settings, progress, and finished chapters in editable Markdown files, so authors can inspect or revise them directly. It supports step-by-step approval or a hands-off mode; personal use is free, while commercial use requires separate permission.

Swift Testing Pro is an agent skill that helps Claude Code, Codex, Cursor, Gemini, and similar assistants write and review tests using Swift Testing. It covers @Test, #expect, #require, parameterized tests, traits, exit tests, confirmations, and other areas where language models often produce incomplete or outdated code. Once installed, it can be triggered by command or natural language and focused on a specific goal, such as parameterizing a suite or adding exit tests. It is intended to improve testing work inside an existing Swift project.

qiaomu-design is a Claude Code skill that begins with a product and interface diagnosis, then creates four deliberately different browser-viewable directions for comparison. After a user selects one, it implements the design using either the existing product system or a newly defined combination of typography, color, layout, motion, and density. It requires Claude Code, and the project recommends Claude or GLM 5.2 when visual quality is the priority.