MCP App StoreMCP App StoreCitadel Setup guides a developer from an empty checkout to a fully configured Citadel harness. It installs lifecycle hooks, auto-detects language/framework/package manager/test tooling, writes a protected harness.json with registered skills, and optionally runs a short live demo and full tour to show core capabilities. The goal is to have hooks active before any interactive steps so sessions are protected from the start.
Run this on a project when you first add Citadel or when you want to rebuild harness configuration. Use Recommended mode for a guided setup, Full Tour to see demonstrations and walkthroughs, or Express for a silent, zero-question install. Also use Update when reconfiguring an existing harness.
Primarily designed for Claude Code harnesses and Claude-powered agents (Claude Code/Claude variants). The setup produces standard configuration consumable by orchestration layers and agents that respect harness.json and the /do routing model.
Citadel Setup is a comprehensive first-run wizard for the Citadel agent orchestration harness. It offers three modes (Recommended, Full Tour, Express) and auto-detects project stack, installs hooks, generates config, and optionally runs live demos. No bundled scripts to execute — all logic is described procedurally in SKILL.md with inline node commands. Well-structured with thorough error handling and fringe cases.
No security red flags. The skill modifies local project config files (.claude/settings.json, .claude/harness.json) which is its stated purpose. Telemetry is local-only. The inline node -e script for harness.json generation is functional but dense. Overall a well-crafted setup wizard with good UX flow and comprehensive edge case handling.
Marshal (Session Commander)
A single-session meta-orchestrator that parses a direction, chains existing skills, and executes multi-step tasks end-to-end without unnecessary user prompts.
/research — Focused Investigation
Structured research protocol that converts questions into focused findings with cited sources, confidence levels, and a written research document for follow-up.
Create App — End-to-End Application Generator
Generate a complete application or add features to an existing codebase from a single natural-language description, with tiered workflows, PRD/architecture chec
PRD (Product Requirements Document) Generator
Generates structured Product Requirements Documents from natural-language app descriptions, asking clarifying questions, researching references, and producing m
Do — Unified Intent Router
Route user intents to the cheapest capable handler (skill, marshal, archon, or fleet) using a multi-tier classifier for scope, complexity, persistence, and para
Fleet — Parallel Campaign Orchestrator
Orchestrate parallel campaigns in waves: spawn isolated agents/worktrees, coordinate discoveries, validate results, and merge work — for multi-stream engineerin
Learn — Knowledge Compiler
Compile findings from completed campaigns and evolve cycles into a structured project wiki, extracting patterns, decisions, and quality rules for reuse.
NGINX Integration
Integrate with NGINX via the Membrane CLI to query, manage, and proxy requests to NGINX for configuration, logs, and runtime operations.
Monad Development
Practical developer skill for deploying and verifying smart contracts, configuring Foundry projects, and integrating frontends (viem/wagmi) on the Monad blockch
Starlark Dev
Create, debug, and test Kurtosis Starlark packages — write package structure, run dry-runs, and inspect plan execution for reliable orchestration.
Maestro Orchestrate
Orchestrates complex engineering workflows with enforced design and approval gates, child-agent delegation, and session state management.
Run Execute
Orchestrates execution of work items across modes (autopilot, confirm, validate) with scripted init/complete tooling, plan/test/report artifacts and strict gati
Uloop: Execute Dynamic Code
Run small C# snippets in the Unity Editor via the uloop CLI for editor automation tasks like prefab wiring, AddComponent flows, and scene edits—without writing
Overnight — Autonomous Long-Running Coding
Orchestrates long-running coding goals: decomposes objectives into atomic tasks, dispatches isolated worktree workers, verifies acceptance criteria, and merges
tmux Remote Control
Control tmux sessions for interactive TTYs: send keystrokes, capture panes, orchestrate parallel coding agents, and monitor output safely.
