CrofAI for GitHub Copilot Chat
Use CrofAI models directly from the GitHub Copilot Chat model picker in Visual Studio Code.
This extension is a native VS Code LanguageModelChatProvider. It validates a user-supplied CrofAI API key, discovers the models available to that key, and streams OpenAI-compatible chat completions directly from https://crof.ai/v1 into Copilot Chat.
Highlights
- Direct CrofAI integration without a local proxy
- API keys managed by VS Code Secret Storage or provider configuration
- Multiple isolated CrofAI API-key entries in Manage Language Models
- Current input, cached-input, and output pricing in the model picker
- Refreshable CrofAI credit balance, daily request allowance, and locally tracked token activity
- Live
/models discovery with context and output limits from CrofAI metadata
- Streaming text,
reasoning_content, token usage, and function-tool calls
- CrofAI reasoning effort controls only on models that advertise support
- Image input for models that advertise image capability
- Agent mode function-tool calls
Quick start
- Install the extension. You need VS Code 1.125 or newer and GitHub Copilot Chat.
- Create an API key in the CrofAI dashboard.
- Open Copilot Chat, select Manage Models, add a CrofAI provider entry, and enter the key.
- Choose any model returned by your CrofAI account.
To use more than one account or key, add another CrofAI entry. Each entry keeps its own credential and model list. The legacy CrofAI: Configure API Key command remains available for command-driven workflows.
For models that support configurable reasoning, choose None, Low, Medium, or High from the reasoning control in Copilot Chat. A per-request selection overrides crofCopilot.reasoningEffort.
Documentation
Unofficial project; not affiliated with CrofAI, GitHub, or Microsoft. CrofAI usage limits and charges still apply. Licensed under MIT.