dev-tools-plugin

Spec-driven development in small, checkable files, where the one who writes a claim never writes its proof. A raw idea is shaped into spec.md in rounds, with sized holes and research in parallel subagents, until the person says done; a big design becomes a roadmap of planned steps, shaped one at a time as their dependencies are built. The spec becomes EARS requirement records, the records become tasks sized for one agent run, and a test-analyst subagent writes the proof plan (tests.yaml) for every scenario and every task: scripts a driver runs, or sessions a tester performs in a browser (the bundled Playwright MCP) or against an API. Test developers write the scripts red before any code; developers build; a bundled script — not the model — gates every stage, decides when a task is done, verifies the whole feature at the tree that ships, and merges through the forge CLI. A failing test is judged by a decider (test, code or spec?) and fixed by the agent that owns it, never by the person; no task and no test may name a person as an actor. Every skill is user-invoked only.

12 skills · 7 agents · 1 MCP server

Media generation and manipulation for Claude Code — images, videos/GIFs, music, text-to-speech, and data visualizations (charts, graphs, diagrams, maps) using Google Gemini, D3.js, Mermaid, and Draw.io. A single visual-planning gate plus media-prompt-craft ensure every asset is clarified, styled, and reviewed before it's produced.

10 skills · 2 commands · 1 agent · 5 MCP servers · hooks

Turns a project repo into a narrated, edited demo video. Reads the codebase, writes a storyboard with real narrative value, prepares deterministic demo state, drives the UI with Playwright (web, launched Electron, or an attached running app) recording one clip per section with Playwright video or OBS over obs-websocket, generates ElevenLabs voiceover, reconciles measured durations into a timeline, and renders the final cut with Remotion or exports it as a Final Cut Pro project dressed with Motion templates, MotionVFX DesignStudio elements first-class.

9 skills · 5 agents · 2 MCP servers · hooks

End-to-end website/webapp design and implementation — from brief to working React/Vite code. Orchestrates design direction, content architecture, media generation, parallel per-page implementation, and visual testing with an opinionated anti-slop workflow.

4 skills · 1 command · 11 agents · 1 MCP server

Design direction and creative guidance for Claude Code — styleguides, aesthetic strategy, typography pairings, color mood systems, and design review. The 'taste layer' that makes AI-assisted design intentional rather than generic. Works with media-plugin (which now owns media prompt crafting) for asset generation.

4 skills · 1 command · 1 agent

Infrastructure management for Kubernetes/GKE, Istio service mesh, and authentication (Keycloak, OAuth2-proxy). Includes Helm charts, Terraform IaC, and Traefik reverse proxy.

6 skills

Recommends relevant Claude Academy courses, tutorials, and use cases when users ask how to use Claude

19 skills

AGENTS.md as project instructions, by one option. Under claude-md-or-agents-md a project with no instruction files of its own gets its AGENTS.md files instead: a prompt.context hook asks $.fs.ancestors for every AGENTS.md and .claude/AGENTS.md from the filesystem root down to the working directory and hands them to the engine as project instruction files, which it renders, announces and withholds from agents exactly as it does CLAUDE.md, skipping any file the engine already loaded by import or link; a tool.call hook on Read attaches the ones under the project root as the engine attaches a nested CLAUDE.md. Under claude-md-and-agents-md every AGENTS.md is loaded beside CLAUDE.md. Under managed-only the project's and the person's instruction files are dropped and the organization's managed ones kept. Under claude-md the engine's walk stands alone. claude-md-or-agents-md is the default.

AI-powered codebase understanding — analyze, visualize, and explain any project

9 skills · 10 agents · hooks

DataRobot skills for AI/ML workflows — model training, deployment, predictions, feature engineering, monitoring, explainability, data preparation, App Framework CI/CD, and external agent monitoring.

19 skills

Independent auditing for AI agents, open source. Claude discovers your setup, confirms the cost, and audits your agent against the one question other tools skip: is it doing the job it's supposed to do, given your business rules and org structure? Test any model (Anthropic, OpenAI, Gemini, Azure, Bedrock, …) graded by self, one independent judge, or a cross-vendor panel; each provider is billed to its own account via a key in your settings, never the command line. Results render as an interactive artifact (with a static-report fallback).

1 skill · hooks

Pre-built small business workflows to help you run and grow your business. Supports a majority of the tools you already use. You approve every step that touches money or customers. Install and then ask Claude to help you get onboarded.

44 skills · 35 MCP servers