am-will/codex-skills▌

Multi-agent planning council that orchestrates independent implementation plans, anonymizes them, then merges into one final plan. \n \n Supports configurable planner agents (Codex, Claude, Gemini, OpenCode, or custom CLI commands) running in parallel, with optional judge override \n Conducts structured intake questioning before plan generation to clarify ambiguities, constraints, and success criteria \n Produces validated Markdown outputs with automatic retry logic (up to 2 attempts) and failur

Build distinctive, production-grade frontend interfaces that avoid generic AI aesthetics. \n \n Guides design thinking through purpose, tone, and differentiation before implementation, emphasizing bold aesthetic direction over generic defaults \n Covers five core design areas: typography (distinctive font pairing), color and theming (CSS variables, dominant colors with sharp accents), motion (CSS animations, scroll triggers, staggered reveals), spatial composition (asymmetry, overlap, grid-break

Spawn autonomous subagents to offload context-heavy work and preserve parent token budget. \n \n Subagents burn their own tokens and return only final results, ideal for deep research (3+ searches), codebase exploration (8+ files), multi-step workflows, and long-running operations \n Choose between mini model (gpt-5.1-codex-mini) for pure search tasks or inherit parent model for multi-step analysis, refactoring, and generation work \n Supports up to 5 parallel subagents via background shell exec

Query OpenAI's official developer documentation via MCP server for API, SDK, and platform guidance. \n \n Search and fetch docs on OpenAI APIs (Chat Completions, Realtime, Responses), SDKs, ChatGPT Apps, Codex, and migrations using a CLI wrapper around the OpenAI Docs MCP server \n Includes subcommands for discovery ( search , list ), retrieval ( fetch ), and schema inspection ( endpoints , openapi ) \n Supports optional anchors for fetching specific doc sections and code sample generation in mu

$22

Convert any website into clean, extractable Markdown by routing through markdown.new. \n \n Automatically rewrites URLs to prefix them with https://markdown.new/ for cleaner text extraction from docs, blog posts, changelogs, and GitHub issues \n Includes a rewrite rule that preserves query parameters, fragments, and schemes while maintaining the original URL structure \n Designed for read-only content only; skips markdown.new for authenticated workflows, interactive dashboards, paywalls, and sit

Orchestrates parallel task execution across up to 12 concurrent Sparky subagents using a rolling pool scheduler. \n \n Parses markdown plan files, extracts task definitions, and launches subagents continuously as slots open without waiting for batch completion \n Maintains canonical file paths and naming constraints across parallel tasks to prevent filename drift and cross-task conflicts \n Validates each subagent result, updates the plan file with completion logs, and immediately schedules the

Fetch current library documentation via Context7 API for any external package. \n \n Search for libraries by name, then retrieve documentation context using the library ID and a specific query \n Supports two output formats (txt and markdown) with configurable token limits for response size \n Use proactively when working with external libraries, installing dependencies, debugging library-specific issues, or verifying APIs beyond your training data cutoff \n API key stored in .env file within th

Writes failing tests first, establishing a strict implementation contract for test-driven development. \n \n Completes the RED phase of TDD by authoring behavior-focused tests that fail for the right reason, then hands off to implementation agents with objective pass criteria \n Delegates test authoring to a tdd_test_writer subagent when available, ensuring tests remain unmodified during implementation \n Detects existing test frameworks and conventions (pytest, jest, vitest, go test, cargo test

Orchestrate parallel development tasks with dependency management and test-driven validation. \n \n Parses markdown plan files to extract task definitions, dependencies, and acceptance criteria, then launches unblocked tasks in parallel waves using Sparky subagents \n Enforces test-driven development (RED phase first) for testable tasks, with fallback to documented non-testable verification (manual, static, or runtime checks) \n Manages task dependencies automatically, blocking tasks until their

Orchestrates parallel execution of dependent tasks across multiple subagents using markdown plan files. \n \n Parses task dependencies from plan markdown and launches unblocked tasks in parallel waves, ensuring correct execution order \n Provides subagents with structured task context including descriptions, acceptance criteria, validation steps, and dependency information \n Enforces test-driven development (RED phase first) with fallback to documented non-testable verification for tasks that d

Read and search GitHub repository documentation through the gitmcp.io MCP service. \n \n Converts GitHub URLs to gitmcp.io endpoints and provides CLI access via scripts/gitmcp.py for fetching docs, searching documentation semantically, and searching code via GitHub API \n Four core tools available per repository: fetch full documentation, semantic search within docs, exact-match code search, and fetch external URLs referenced in documentation \n Tool names are dynamically generated and prefixed

Fast Rust-based headless browser automation with real Chrome profile and session support. \n \n Supports headless Chromium, real Chrome with profile/OAuth reuse via Chrome DevTools Protocol, and persistent session cookies for logged-in workflows \n 30+ commands covering navigation, interaction (click, type, fill, drag, upload), extraction (text, HTML, attributes, screenshots), and state checks \n Deterministic element selection via snapshot refs (@e1, @e2, etc.) and semantic finders (role, text,

Detailed, phased implementation plans with sprints and atomic tasks for bugs, features, or tasks. \n \n Structures plans into research, requirements clarification, sprint-based implementation, and risk assessment phases \n Breaks work into demoable, independently testable sprints with specific, committable atomic tasks including file paths and dependencies \n Asks targeted clarification questions upfront to resolve ambiguities around scope, constraints, priorities, and success criteria \n Identi

45 performance optimization rules for React and Next.js, organized by impact and category. \n \n Covers 8 rule categories from eliminating waterfalls and bundle optimization (critical) through JavaScript performance and advanced patterns (low priority) \n Each rule includes explanation, before/after code examples, and context to guide automated refactoring and code generation \n Prioritized by impact: waterfalls and bundle size are critical; server-side performance and re-render optimization are

Gemini 2.5 Computer Use browser automation with Playwright-based agent loops and safety confirmations. \n \n Implements a screenshot-to-action cycle: capture screen, send to Gemini, parse function calls, execute in Playwright, return results until task completion or turn limit \n Supports multiple browser options: bundled Chromium (default), Chrome/Edge channels via COMPUTER_USE_BROWSER_CHANNEL , or custom executables like Brave \n Includes safety confirmation workflow that prompts users before

Generates dependency-aware implementation plans optimized for parallel multi-agent execution. \n \n Creates atomic, independently executable tasks with explicit dependency declarations to maximize parallelization across agents \n Requires codebase investigation, documentation retrieval for external libraries, and clarifying questions before planning to eliminate ambiguity \n Structures plans with task IDs, dependency arrays, descriptions, file locations, and validation criteria; includes visual

Create and install custom Codex agent roles with guided configuration and validation. \n \n Collects required inputs (model, reasoning effort, developer instructions, role name, install scope) before writing any files, with strict validation against the config schema \n Writes minimal role config files by default, adding optional parameters like sandboxing, web search controls, and MCP servers only when explicitly requested \n Installs roles into global ( ~/.codex/config.toml ) or project-scoped

Comprehensive, phased implementation plans broken into sprints and atomic tasks. \n \n Guides you through requirement clarification before planning, asking 5-10 targeted questions to reduce ambiguity and surface risks \n Fetches official documentation for external libraries, APIs, and frameworks to ensure version-accurate steps and current best practices \n Structures plans into logical sprints that each produce demoable, testable increments, with atomic, independently committable tasks includin