kimi-code

name	kimi-code
description	Terminal AI coding agent with video input, subagents, MCP support, and ACP editor integration
version	1.0.0
author	Hermes Agent
license	MIT
platforms	["linux","macos","windows"]
metadata	{"hermes":{"tags":["ai-agent","coding","cli","multimodal","mcp"],"related_skills":["claude-code","opencode"]}}

Kimi Code — Terminal AI Coding Agent

A terminal-based AI coding agent from Moonshot AI with video input support, sub-agent orchestration, MCP tool integration, and ACP editor connections. Ships as a single binary.

Prerequisites

• Node.js 18+ or standalone binary
• LLM API key (configurable provider)

Installation

# Install via npm
npm install -g @moonshotai/kimi-code

# Or download binary from GitHub releases
# https://github.com/MoonshotAI/kimi-code/releases

Usage

Basic Coding

# Start interactive session
kimi-code

# Single-shot coding task
kimi-code "add error handling to the auth module"

# Use video input for visual context
kimi-code --video screenshot.png "fix this UI bug"

MCP Integration

# Configure MCP servers
kimi-code --mcp-server filesystem --mcp-server github

# Use MCP tools in coding sessions
kimi-code "use the github tool to create a PR for these changes"

Sub-agent Orchestration

# Spawn sub-agents for parallel tasks
kimi-code "refactor the database layer and write tests in parallel"

ACP Editor Connection

# Connect to VS Code via ACP
kimi-code --acp-editor vscode

# Connect to Cursor via ACP
kimi-code --acp-editor cursor

Common Pitfalls

• Video input size: Large screenshots may exceed context limits. Resize before passing.
• MCP server availability: Ensure MCP servers are running before starting kimi-code.
• Sub-agent costs: Each sub-agent consumes separate LLM tokens.

Verification

# Verify installation
kimi-code --version

# Test basic functionality
echo "print('hello')" > test.py
kimi-code "add a docstring to this function" test.py
cat test.py