magus

Universal development assistant. Detects the project stack and routes work to specialist agents for architecture, implementation, debugging, review and research, enforcing design-system guardrails on every UI change.

19 skills · 17 commands · 12 agents · hooks

Bun and TypeScript toolkit. A `/bunjs:bun` index routes any task to the right one of eight task-shaped skills — setup, HTTP, errors, testing, security, production, performance, TUI — each shipping copyable code verified against Bun. Only Bun is tested.

9 skills · 9 commands

Slot-addressed terminal control over tmux or herdr. Runs interactive commands, dev servers, test watchers, REPLs and TUI apps in numbered helper panes — beside the user or isolated from view — that the server places and owns, so an agent never holds a pane id or targets the user's own session.

5 skills · 9 commands · 1 agent · 1 MCP server · hooks

UI design validation. Compares a rendered screen against its reference by pixel diff, then reviews the result for spacing, hierarchy and design-system consistency.

6 skills · 3 commands · 2 agents

Read-only code search and investigation behind one stable tool surface. Finds code by meaning or structure, traces callers, callees and blast radius, and dispatches an analyze agent for deep dives. The search engine is named in project settings and is swappable.

4 skills · 3 commands · 1 agent · 1 MCP server · hooks

Full-platform browser automation through its own MCP server: navigation, clicks and typing, live-page JS eval, keyboard and focus, file screenshots, session export and a configurable agent LLM, on local or cloud browsers. Runs Playwright's own Chromium, never your real Chrome. Ships 6 skills.

6 skills · 1 MCP server · hooks

Project setup jobs in one plugin. Investigates a repository and provisions it — plugins, tools, framework best-practice references, and a seeded knowledge base — then installs the adaptive statusline and indexes every skill reachable from the project into browsable markdown.

2 skills · 4 commands · hooks

Toolkit for madbench, MadAppGang's Go harness for benchmarking agentic coding tools. Carries madbench's own skills with /madbench:madbench as the entry point, checks them against the installed madbench, and runs benches in a visible pane through the madbench:bench agent.

3 skills · 1 command · 1 agent · hooks

Multi-model collaboration and orchestration. Runs one task across several AI models under a task type (judge, research, plan, debug, brainstorm, implement, test, collaborate), each with its own result, or delegates it to one model running a full Claude Code session.

2 skills · 2 commands · 1 agent · hooks

Runs ticket work as three levels of Claude Code sessions: triage owns tickets and the plan, a manager per ticket works in its own tmux window or herdr tab and worktree, children in panes beside it. A status band lists them; a guard keeps ticket changes with triage. Needs bun, git, tmux or herdr.

3 skills · 1 command · hooks

Skills for Claude Code in a meroku project: troubleshoot a failed meroku operation from its Fix with AI briefing, change an environment YAML safely, place new AWS infrastructure beside meroku's own, and understand how the YAML becomes Terraform. Investigates read-only and asks before any write.

4 skills

MCP runtime for external AI models. Proxies OpenRouter, Ollama and LM Studio, runs blind team voting, manages async sessions, and shows the session's runs live above the prompt, with stop and live-screen tabs. Required by code-search, dev, multimodel and designer.

1 skill · 1 MCP server · hooks

Hands browser tasks to Aside, the user's own AI browser with its own logged-in agent, through the aside CLI, in a split terminal pane when the terminal plugin is present. An alternative to browser-use for work in the user's accounts. Needs the aside CLI.

1 skill · 1 command

MCP runtime for semantic code search and AST analysis, ranking symbols by PageRank. Optional engine behind the code-search facade, and usable directly for its full tool surface.

1 MCP server · hooks

BETA — Dingo language toolkit. Idiomatic Result and Option types, error propagation, enums, pattern matching, lambdas and safe navigation, with compiler-verified patterns. Interfaces may still change.

1 skill

Go development toolkit. Ships the go-tui skill for building Charm-stack terminal UIs (Bubble Tea, Lip Gloss, Bubbles, ntcharts), with a colour-accurate screenshot workflow for reviewing a running TUI.

1 skill