Repl Toolkit

A .NET framework for building composable command surfaces.

• Define your commands once — run them as a CLI, explore them in an interactive REPL,
• host them in session-based terminals, expose them as MCP servers for AI agents,
• or drive them from automation scripts.

New here? The DeepWiki has full architecture docs, diagrams, and an AI assistant you can ask questions about the toolkit.

Quick start

dotnet add package Repl

using Repl;

var app = ReplApp.Create().UseDefaultInteractive();
app.Map("hello", () => "world");
return app.Run(args);

Example

using Repl;

var app = ReplApp.Create().UseDefaultInteractive();

app.Context("client", client =>
{
    client.Map("list", () => new { Clients = new[] { "ACME", "Globex" } });

    client.Context("{id:int}", scoped =>
    {
        scoped.Map("show", (int id) => new { Id = id, Name = "ACME" });
        scoped.Map("remove", (int id) => Results.Cancelled($"Remove {id} cancelled."));
    });
});

return app.Run(args);

CLI mode:

$ myapp client list --json
{
  "clients": ["ACME", "Globex"]
}

REPL mode (same command graph):

$ myapp
> client 42 show --json
{ "id": 42, "name": "ACME" }

> client
[client]> list
ACME
Globex

MCP mode (same command graph, exposed to AI agents):

using Repl.Mcp;

app.UseMcpServer();  // add one line

{ "command": "myapp", "args": ["mcp", "serve"] }

One command graph. CLI, REPL, remote sessions, and AI agents — all from the same code.

What's included

Feature	Package	Guides
Unified command graph — routing, constraints, binding		Route system Execution pipeline Commands
Interactive REPL — scopes, history, autocomplete		Interactive loop Configuration
Parameters & options — typed binding, options groups, response files		Parameter system Route system
Multiple output formats — JSON, XML, YAML, Markdown		Output system
MCP server — expose commands as AI agent tools		MCP server MCP advanced
Typed results & interactions — prompts, progress, cancellation		Interaction channel
Session hosting — WebSocket, Telnet, remote terminals		Runtime channels Terminal metadata
Shell completion — Bash, PowerShell, Zsh, Fish, Nushell		Shell completion
Spectre.Console — rich prompts, tables, charts		Interaction channel Sample
Testing toolkit — in-memory multi-session harness		Testing toolkit
Machine-readable contracts — help schemas, error contracts		Help system
Conditional modules — channel-aware, feature-gated commands		Module presence

Repl is the meta-package that bundles Core + Defaults + Protocol — start here.

Learn by example

Progressive learning path — each sample builds on the previous:

1. Core Basics — routing, constraints, help, output modes
2. Scoped Contacts — dynamic scoping, .. navigation
3. Modular Ops — composable modules, generic CRUD
4. Interactive Ops — prompts, progress, timeouts, cancellation
5. Hosting Remote — WebSocket / Telnet session hosting
6. Testing — multi-session typed assertions
7. Spectre — Spectre.Console renderables, visualizations, rich prompts
8. MCP Server — expose commands as MCP tools for AI agents

More documentation

Architecture blueprint	Best practices
Comparison & migration	Publishing & deployment
Glossary	Interactive docs & AI Q&A

AI-assisted development

If you use an AI coding agent (Claude Code, Cursor, Windsurf, Copilot, Codex, OpenCode, etc.) to build your Repl Toolkit app, install Context7 to give it access to Repl Toolkit documentation and code snippets.

Library ID: /yllibed/repl

Most agents that support MCP can use Context7 directly. Refer to your agent's documentation for MCP server configuration.

You can also explore the toolkit architecture and ask questions on DeepWiki.

Contributing

Contributions welcome — please discuss new features first to keep the toolkit aligned with its goals. See CONTRIBUTING.md, CODE_OF_CONDUCT.md, and SECURITY.md.

License

MIT — Copyright (c) 2026 Yllibed project / Carl de Billy