LiuMengxuan04

MiniCode Architecture

Development community intermediate

SOURCE https://github.com/LiuMengxuan04/MiniCode

Description

[简体中文](./ARCHITECTURE_ZH.md) This document describes the lightweight architecture decisions behind `mini-code`. The goal is not to build a giant all-in-one terminal agent platform, but to prioritize t

Installation

This entry records only its repository, not the path inside it, so there is no exact command to give. Open the source below and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

Repository README

This is the README for LiuMengxuan04/MiniCode, shared by 5 entries in this directory. It describes the repository, not this entry specifically.

MiniCode Architecture

[简体中文](./ARCHITECTURE_ZH.md)

This document describes the lightweight architecture decisions behind `mini-code`. The goal is not to build a giant all-in-one terminal agent platform, but to prioritize the most valuable execution loop, interaction experience, and safety boundaries.

Design Principles

MiniCode prioritizes these capabilities:

the main model -> tool -> model loop
full-screen TUI interaction rhythm
directory awareness, permission checks, and dangerous-action confirmation
a componentized transcript / tool / input UI structure
a user-reviewable file modification flow

In other words, MiniCode is a smaller, more controllable terminal coding assistant.

Current implementation focus

Keep the skeleton of the model -> tool -> model loop
Keep a unified tool contract and centralized registration
Keep a message-driven terminal interaction rhythm
Keep safety boundaries: path permissions, command permissions, and write approval
Keep Claude Code-inspired extension points: local skills and MCP-backed tools

Planned / not yet built

Full Ink/React rendering stack
Bridge / IDE two-way communication
Remote session
Task swarm / sub-agent orchestration
LSP
Skill marketplace
More complex permission modes
Feature-flag system
Telemetry / analytics
Compact / memory / session restore

Current implementation

src/index.ts: CLI entry
src/agent-loop.ts: multi-turn tool-calling loop
src/tool.ts: registration, validation, execution
src/tools/*: list_files / grep_files / read_file / write_file / edit_file / patch_file / modify_file / run_command / web_fetch / web_search / ask_user / load_skill
src/config.ts: uses dedicated ~/.mini-code
src/skills.ts: scans .mini-code/skills and compatible .claude/skills directories
src/mcp.ts: launches stdio MCP servers, negotiates framing compatibility, and wraps remote MCP tools into local tool definitions
src/background-tasks.ts: minimal background shell task registry used by run_command and the TUI
src/manage-cli.ts: manages persisted MCP configs and installed local skills
src/anthropic-adapter.ts: Anthropic-compatible Messages API adapter
src/mock-model.ts: offline fallback adapter
src/permissions.ts: path, command, and edit approval with allowlist / denylist
src/file-review.ts: diff review before writing files
src/tui/*: transcript / chrome / input / screen / markdown terminal components

Why it is good for learning

One strength of MiniCode is that it delivers Claude Code–like behavior and core architectural ideas in a much lighter implementation.

That makes it well suited to:

Learning the basic pieces of a terminal coding agent
Studying tool-calling loops
Understanding permission approval and file review flows
Seeing how skills and external MCP tools can be added without a heavy plugin platform
Seeing a lightweight Claude Code-style di