> For the complete documentation index, see [llms.txt](https://docs.collieai.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.collieai.io/proxy-integration.md).

# Proxy Integration

- [Proxy integration overview](https://docs.collieai.io/proxy-integration/overview.md): How the CollieAi proxy integration works — a drop-in, provider-agnostic proxy that filters every LLM request and response, OpenAI-compatible and Anthropic-native.
- [OpenAI SDK Integration](https://docs.collieai.io/proxy-integration/openai-sdk-integration.md): Connect your app to CollieAi with the OpenAI SDK — Python, Node.js, cURL, and function calling. Point the OpenAI client at CollieAi by changing the base\_url.
- [Anthropic SDK Integration](https://docs.collieai.io/proxy-integration/anthropic-sdk-integration.md): Connect Anthropic's Claude to CollieAi — use the native Messages API (/v1/messages) or the OpenAI-compatible endpoint with claude-\* models, secured by the same guardrails.
- [Streaming](https://docs.collieai.io/proxy-integration/streaming.md): How CollieAi streams LLM responses — real-time SSE streaming with output guardrails enforced on the stream, in both incremental and buffered modes, for OpenAI and Anthropic.
- [Error handling](https://docs.collieai.io/proxy-integration/error-handling.md): How CollieAi reports errors — the OpenAI-compatible error format, error types, policy-violation responses with triggered rules, and troubleshooting steps.