Shape agent capabilities into your workflow.

Decision Memory watches product conversations for moments such as accepting, rejecting, deferring, merging, or shipping an idea. It asks for the reasoning and any revisit or notification details, then formats the result as a searchable decision record and closure-message draft. The skill works on its own with Markdown output, while an optional IdeaLift MCP connection can persist records and link them to ideas.

Use this skill when creating Gutenberg blocks, dynamic rendering, or Full Site Editing features in Claude Code. It covers block.json, InnerBlocks, variations, transforms, WordPress scripts, and server-side rendering, and can activate automatically or be invoked with /wp-gutenberg.

Use this skill to review or implement accessibility in WordPress themes with Claude Code. Its guidance covers WCAG 2.1 AA concerns such as skip links, landmarks, color contrast, forms, screen-reader support, keyboard navigation, and focus management, with automatic or direct invocation after installation.

Use this Claude Code skill when moving a WordPress plugin or theme to PHP 8.x. It covers language features such as named arguments, union types, match expressions, enums, readonly properties, and constructor promotion while keeping backward-compatibility strategies in view.

Council sends the same review prompt to several configured AI coding agents in parallel, then asks the host agent to synthesize a deduplicated report for staged Git changes. It can mix Claude Code, Codex, OpenCode, and Cursor backends, but requires at least one supported and authenticated CLI.

Zero-Context Validation checks whether documentation is truly sufficient by giving a fresh sub-agent only the docs and a concrete implementation task. Any missing interface, file location, or contract must be reported instead of guessed, so the main agent can revise the material and repeat the test. It is designed for Claude Code or compatible agents and uses their built-in task mechanism.

UED Copilot reviews screenshots or design-system materials for visual hierarchy, interaction flow, usability, accessibility, and platform conventions across web, iOS, and Android. It can use a team's own standards or fall back to Apple HIG, Material Design, Nielsen heuristics, and WCAG guidance. The Claude Code skill presents issues interactively and can include concrete code changes and implementation steps.

OpenClaw Auditor gives Claude Code a static review workflow for proposed OpenClaw configuration changes. It is designed to consult the current documentation index, capture a configuration snapshot, compare schema expectations, score risk, and relate findings to documented failure modes before anything is applied. The skill can format results for Notion and requires curl and jq, with the OpenClaw CLI optional.

hwpx-skill equips Claude Code with workflows for generating HWPX documents from Markdown or text, extracting content, replacing template fields, filling tables, cloning forms, editing existing files, and merging documents. Its tools work directly with the HWPX ZIP and XML structure and include checks for layout drift and unsafe archives. Running the included scripts requires Python and supporting parsing libraries.

Edge TTS Skill lets Claude or a standalone Python command turn typed text and text files into audio through Microsoft Edge's online speech service. Users can select voices in Chinese and other languages and adjust speaking rate and volume. It can be installed as an Agent Skill or run directly from the included synthesizer script, which requires network access and Python dependencies.

token-saver gives Claude Code task-based model guidance, reserving lighter models for extraction and routine work while escalating demanding debugging or multi-file execution. It also promotes one task per conversation, sharing only relevant excerpts, and storing reusable context in CLAUDE.md, memory files, or skills. The project is advisory guidance rather than a separate billing, proxy, or model-routing service.

GitHub Browser Skill teaches Claude Code how to map GitHub URLs and research goals to focused gh CLI and jq commands. It covers repository trees and files, issue and pull-request context, diffs, review comments, code search, and multi-channel repository discovery. The documented workflows are read-only and require an installed, authenticated GitHub CLI.

data-arch-guard helps Claude Code evaluate storage decisions before they turn into maintenance debt. It uses a decision tree and checklist to distinguish database, JSONL, and JSON use cases, and includes a Bash scanner for duplicate formats, bridge scripts, and oversized data files.

OMNX Code sets up and maintains a living CLAUDE.md for React, TypeScript, Supabase, and Vercel projects, installs a security-auditor skill through a verified update flow, and keeps tasks visible. It works with Claude Code or Codex and asks for explicit confirmation before security fixes or other risk-sensitive changes.

Image Prompt Generator turns a visual brief into three concepts, delegates style and booru reference research, then writes final prompts and runs parallel image generation and review. It is built around Codex image generation, with Python and gallery-dl for reference collection, while also adapting prompts for Midjourney, Stable Diffusion, Flux, and other tools.

Formant turns a conversational brief into a form schema and a single self-contained HTML file with one-question-at-a-time navigation, keyboard controls, and light or dark mode. Forms can run offline or deploy to Cloudflare or Vercel. Add a backend only when you need stored responses, analytics, exports, or shared administration.

Pubky Stack Skill gives Claude focused references for the Pubky SDK, Pkarr identity and routing, Pubky Ring authentication, homeserver data, app schemas, and indexer patterns. It includes reusable templates for authentication, typed CRUD, Next.js setup, and custom indexers, making it useful for developers building Pubky-based applications.

Akanuke detects recurring LLM writing habits in English and Japanese, including inflated claims, promotional tone, excessive hedging, rigid formatting, monotonous rhythm, and clichéd metaphors, then rewrites the text in general, formal, academic, or medical modes. An optional report explains the detected patterns; the result should still be checked for preserved meaning and facts.