From f7aac0785b8f42f90cca17e7b60f1c8d3562d304 Mon Sep 17 00:00:00 2001
From: vikingowl <christian@nachtigall.dev>
Date: Fri, 26 Dec 2025 18:30:05 +0100
Subject: [PATCH] docs(llm): Add README.md for all LLM crates

---
 crates/llm/anthropic/README.md | 15 +++++++++++++++
 crates/llm/core/README.md      | 18 ++++++++++++++++++
 crates/llm/ollama/README.md    | 14 ++++++++++++++
 crates/llm/openai/README.md    | 14 ++++++++++++++
 4 files changed, 61 insertions(+)
 create mode 100644 crates/llm/anthropic/README.md
 create mode 100644 crates/llm/core/README.md
 create mode 100644 crates/llm/ollama/README.md
 create mode 100644 crates/llm/openai/README.md

diff --git a/crates/llm/anthropic/README.md b/crates/llm/anthropic/README.md
new file mode 100644
index 0000000..16f7dde
--- /dev/null
+++ b/crates/llm/anthropic/README.md
@@ -0,0 +1,15 @@
+# Owlen Anthropic Provider
+
+Anthropic Claude integration for the Owlen AI agent.
+
+## Overview
+This crate provides the implementation of the `LlmProvider` trait for Anthropic's Claude models. It handles the specific API requirements for Claude, including its unique tool calling format and streaming response structure.
+
+## Features
+- **Claude 3.5 Sonnet Support:** Optimized for the latest high-performance models.
+- **Tool Use:** Native integration with Claude's tool calling capabilities.
+- **Streaming:** Efficient real-time response generation using server-sent events.
+- **Token Counting:** Accurate token estimation using Anthropic-specific logic.
+
+## Configuration
+Requires an `ANTHROPIC_API_KEY` to be set in the environment or configuration.
diff --git a/crates/llm/core/README.md b/crates/llm/core/README.md
new file mode 100644
index 0000000..be9f406
--- /dev/null
+++ b/crates/llm/core/README.md
@@ -0,0 +1,18 @@
+# Owlen LLM Core
+
+The core abstraction layer for Large Language Model (LLM) providers in the Owlen AI agent.
+
+## Overview
+This crate defines the common traits and types used to integrate various AI providers. It enables the agent to be model-agnostic and switch between different backends at runtime.
+
+## Key Components
+- `LlmProvider` Trait: The primary interface that all provider implementations (Anthropic, OpenAI, Ollama) must satisfy.
+- `ChatMessage`: Unified message structure for conversation history.
+- `StreamChunk`: Standardized format for streaming LLM responses.
+- `ToolCall`: Abstraction for tool invocation requests from the model.
+- `TokenCounter`: Utilities for estimating token usage and managing context windows.
+
+## Supported Providers
+- **Anthropic:** Integration with Claude 3.5 models.
+- **OpenAI:** Integration with GPT-4o models.
+- **Ollama:** Support for local models (e.g., Llama 3, Qwen).
diff --git a/crates/llm/ollama/README.md b/crates/llm/ollama/README.md
new file mode 100644
index 0000000..ca6a385
--- /dev/null
+++ b/crates/llm/ollama/README.md
@@ -0,0 +1,14 @@
+# Owlen Ollama Provider
+
+Local LLM integration via Ollama for the Owlen AI agent.
+
+## Overview
+This crate enables the Owlen agent to use local models running via Ollama. This is ideal for privacy-focused workflows or development without an internet connection.
+
+## Features
+- **Local Execution:** No API keys required for basic local use.
+- **Llama 3 / Qwen Support:** Compatible with popular open-source models.
+- **Custom Model URLs:** Connect to Ollama instances running on non-standard ports or remote servers.
+
+## Configuration
+Requires a running Ollama instance. The default connection URL is `http://localhost:11434`.
diff --git a/crates/llm/openai/README.md b/crates/llm/openai/README.md
new file mode 100644
index 0000000..22ef2d8
--- /dev/null
+++ b/crates/llm/openai/README.md
@@ -0,0 +1,14 @@
+# Owlen OpenAI Provider
+
+OpenAI GPT integration for the Owlen AI agent.
+
+## Overview
+This crate provides the implementation of the `LlmProvider` trait for OpenAI's models (e.g., GPT-4o). It supports the standard OpenAI chat completion API, including tool calling and streaming.
+
+## Features
+- **GPT-4o Support:** Reliable integration with flagship OpenAI models.
+- **Function Calling:** Full support for OpenAI's tool/function calling mechanism.
+- **Streaming:** Robust real-time response generation.
+
+## Configuration
+Requires an `OPENAI_API_KEY` to be set in the environment or configuration.