By shawn-sandy
Plan creation and review on demand — /plan-agent:implementation-plan skill runs the full Steps 1-8 workflow with a built-in structured interview; /plan-agent:review-plan skill spawns an Agent Team (5 core + 2 UI-conditional reviewers) to review, improve, and update implementation plans in place; /plan-agent:finalize-plan skill checks completion evidence and sets status; plans-library generates a filterable HTML gallery; filename hook enforces verb-target kebab-case
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Background plan-review agent. Runs the seven-reviewer Agent Team (architecture, completeness, testability, risk, conventions + conditional UX and accessibility) on an implementation plan without blocking the parent session. Improves and updates the plan in place. Use when the user asks to "review in the background", "fire off the review", or "review and improve this plan while I keep working". Mirrors the review-plan skill but runs as a background subagent.
Reviews UI implementation plans for WCAG 2.1 AA compliance, semantic HTML, keyboard navigation, and screen reader support. Use only for plans that touch React, Vue, UI components, or user-facing flows.
Reviews implementation plans for architectural soundness, design pattern fit, and system integration coherence.
Reviews implementation plans for completeness — ensuring no steps are vague, no critical files are omitted, and the path from plan to done is unbroken.
Reviews implementation plans for adherence to project conventions, code style, naming patterns, and organizational consistency.
Generate an HTML implementation-plan document. Produces a self-contained .html plan file with steps, acceptance criteria, and metadata. Use when the user asks to create a plan document, generate an HTML plan, or write a plan file — not for general planning questions.
Builds and opens a filterable HTML gallery of all plans in the plans directory. Scans HTML plans, parses metadata, and writes index.html. Use when asked to browse plans or view plan history.
Craft-prompt: interviews users and assembles a structured AI prompt using Anthropic best-practice techniques. Use when the user runs /plan-agent:craft-prompt or asks to craft a prompt.
Marks an HTML plan as completed. Inspects codebase evidence, confirms with the user, and ticks every acceptance-criteria checkbox. Use via /plan-agent:finalize-plan.
Opens the existing plans gallery without rebuilding. Opens index.html directly; run plans-library first if not yet built. Use when asked to reopen the plans gallery or browse plans without a rebuild.
Uses power tools
Uses Bash, Write, or Edit tools
A marketplace system for Claude Code plugins — enabling discovery, distribution, and installation of AI-powered plugins that extend Claude's capabilities across code review, planning, testing, git workflows, accessibility, and more.
Marketplace: agentics-kit v4.0.0 · 12 plugins · Requires Claude Code 1.0.33+ · View all plugins · Browse docs
Breaking change — v4.0.0: Six plugins have been removed from the marketplace:
agent-creator,agent-reviewer,agentic-plugin-dev,code-simplifier,marketplace-builder, andreact-perf-analyzer. Their source directories are retained in the repository but are no longer installable via the marketplace. See CHANGELOG.md for details.
Install from the marketplace (no cloning required):
# 1. Register the agentics-kit marketplace in Claude Code
/plugin marketplace add shawn-sandy/agentics-kit
# 2. Install the plugins you want
/plugin install code-review@agentics-kit
/plugin install git-agent@agentics-kit
/plugin install memory-tools@agentics-kit
Or load any plugin directly for local testing:
git clone https://github.com/shawn-sandy/agentics.git
cd agentics
claude --plugin-dir ./kit/plugins/code-review
# Then ask: "Review this code for issues"
The agentics project serves two purposes:
| Purpose | What it contains |
|---|---|
| Active Plugins | 12 marketplace plugins in kit/plugins/ — installable via /plugin install, covering code review, planning, testing, git workflows, accessibility, and more |
| Archived Plugins | 6 removed plugins with source directories retained in kit/plugins/ — loadable locally via --plugin-dir but not available via /plugin install |
| Marketplace Infrastructure | agentics-kit marketplace manifest (marketplace.json) that enables installation via /plugin install |
Every plugin in this repo is a working, production-quality tool you can install and use immediately.
# Verify version
claude --version
Install from the official docs if not present.
gh) — used by the git-agent plugin for GitHub PR creationglab) — used by agent-ship for GitLab MR creationjq — JSON processor, useful for debugging plugin manifests| Platform | Support |
|---|---|
| macOS 12.0+ | Full support |
| Linux (Ubuntu 20.04+) | Full support |
| Windows (WSL2) | Recommended over native Windows |
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimnpx claudepluginhub shawn-sandy/agentics-kit --plugin plan-agentReview and plan Claude Code skills, and run tests for changed files — audit SKILL.md files, scaffold new skills, and verify test coverage
Automated git commit and PR creation — stage, commit with conventional messages, and create PRs in one shot
Analyze code and suggest specific, purpose-driven tests tied to actual behavior and intent — not arbitrary coverage
Systematic code review across quality, bugs, security, and best practices
Stress-test implementation plans with structured multi-round interviews before coding begins
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
Claude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.