Tool Integration

1. Supported Tools

Coding Agent Tools

Click and enter the documentation for the tool you want to use below, and follow the corresponding integration guide.

Codex

OpenAI's official CLI tool, a terminal programming assistant supporting OpenAI API-compatible interfaces.

Claude Code

An AI programming assistant by Anthropic that understands codebases, modifies code across files, and supports running commands and tests.

Gemini CLI

Google's official terminal AI assistant tool, supporting the Gemini model series.

OpenCode

An open-source Coding Agent for developers, providing code generation, editing, and task execution capabilities in the terminal.

Kilo Code

A full-workflow intelligent programming tool supporting VS Code, Cursor, JetBrains, and other IDEs.

General Agent Tools

OpenClaw

An open-source AI assistant that runs on your local device, supporting multiple platforms.

2. Coding Endpoints

Coding Flow protocol endpoints for different coding tools:

Protocol	Endpoint URL
OpenAI Chat Completions	https://api.codingflow.ai/v1
Anthropic Messages	https://api.codingflow.ai
Google Gemini	https://api.codingflow.ai

Please select the correct endpoint URL based on the tool you are using. Incorrect endpoint configuration

will prevent you from using the Coding Flow subscription quota.

3. Configuration Examples

General Configuration Steps

1. Find the API model configuration entry in your tool
2. Select the model provider
3. Enter the API Key and select the model

The following example uses Kilo Code to demonstrate how to connect via the OpenAI protocol. Similarly, other tools that support the OpenAI protocol can use the same configuration approach.

Kilo Code

1. Install the Kilo Code Plugin

Using JetBrains as an example, search for the Kilo Code plugin in the plugin marketplace and install it.

2. Configure the Kilo Code Plugin

Select "Bring my own Key".

Fill in the relevant information as follows:

• API Provider: Anthropic
• Base URL: https://api.codingflow.ai
• API Key: Create a new API Key via the Coding Flow Console, in the format sk-apv-cp-xxxxxx.
• Model: Enter a supported model from the Coding Flow documentation, then click "Use custom".

3. Save and Switch Models

After configuration, save the settings and select the Provider you just created on the main page.

4. Start Using

With this setup, you can start using the models for code generation, debugging, task analysis, and more.