Kiro CLI

Provides an MCP interface to manage kiro-cli with multi-session handling, process pooling, and mock mode for development.

python

0

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "vanphappi-kiro-cli-mcp": {
      "command": "uvx",
      "args": [
        "kiro-cli-mcp"
      ],
      "env": {
        "KIRO_MCP_CLI_PATH": "/usr/local/bin/kiro-cli",
        "KIRO_MCP_LOG_LEVEL": "INFO",
        "KIRO_MCP_POOL_SIZE": "5"
      }
    }
  }
}

You run a dedicated MCP server that exposes a stable interface for orchestrating kiro-cli through IDEs and agents. It enables multi-session workspaces, fast response times via a process pool, robust error handling, and a mock mode for development without kiro-cli. This makes it easy to manage multiple projects, switch contexts on the fly, and automate common kiro-cli tasks from your editor or agent workflows.

How to use

You interact with the MCP server through a client that can send chat messages, run kiro-cli commands, manage sessions, and monitor background tasks. Start a dedicated session per project to keep working directories and conversation history isolated. Create a session, then send messages or execute commands to get AI-assisted insights, code reviews, or task automation.

How to install

Prerequisites you need installed on your system before starting:

# Python 3.10 or newer
# Git to clone the repository

From Source: clone the project, install dependencies in editable mode, and start using the MCP server.

git clone https://github.com/your-org/kiro-cli-mcp.git
cd kiro-cli-mcp
pip install -e .

Alternative installation methods are available after publishing to PyPI or via uvx. Use the following commands to install when those options are published.

# After publishing to PyPI
pip install kiro-cli-mcp

# After publishing to uvx registry
uvx install kiro-cli-mcp

Configuration and usage notes

Configure your MCP client in your IDE to point to one or more MCP servers. The provided examples show two standard stdio configurations that you can adapt to your environment.

Available tools

kiro_session_create

Create a new MCP session with optional agent and working directory to isolate project state and conversation context.

kiro_session_list

List all active sessions with their IDs and current state.

kiro_session_switch

Switch the active context to a different session by ID.

kiro_session_end

End a session and clean up its resources.

kiro_session_clear

Clear session history files without removing the session itself.

kiro_session_save

Save the current session state to a persistent file for later restoration.

kiro_chat

Send a chat message to kiro-cli and receive an AI-generated response within the active session.

kiro_command

Execute kiro-cli commands (such as /help or /mcp) within the current session.

kiro_agents_list

List the available custom agents that can be used for responses.

kiro_history

Retrieve the conversation history for the current session.

kiro_history_clear

Clear the conversation history for the active session.

kiro_chat_async

Start a background chat task and monitor progress asynchronously.

kiro_task_status

Poll progress and results for an asynchronous task.

kiro_task_cancel

Cancel a running asynchronous task.

kiro_task_list

List all active asynchronous tasks.

kiro_pool_stats

Get statistics about the process pool and performance.