Renard 🦊

Renard is a developer tool that Logs your LLM interactions across both CLI tools and browsers, giving you a unified view of how you work with AI.

It captures:

• Prompts you type
• Responses from the model
• Session boundaries
• Tool metadata (Claude, OpenAI, Gemini, etc.)

✨ Features

• 🧠 Sync LLM CLI usage (Claude, OpenAI, Gemini)
• 📜 Structured session-based logging
• 🧩 Works with any CLI via renard
• 📂 JSONL logs (easy to query with jq, grep, etc.)
• 💾 Reads conversation history directly from CLI storage (Claude & Gemini)
• 🎯 Accurate conversation capture without terminal parsing

Hook supported LLM CLIs

sudo npm install -g renard
renard install

Supported tools:

claude
openai
gemini

renard will automatically intercept and track their interactions.

How it works:

• Claude & Gemini: Reads conversation history directly from their storage files
  • Claude: ~/.claude/projects/
  • Gemini: ~/.gemini/tmp/*/chats/
• OpenAI: Standard output capture (may be enhanced in future versions)

⚠️ Important Privacy & Data Safety Note

Renard is built with strict data filtering on the backend.

• ✅ Only developer-context data (prompts, responses, tool metadata, and session structure) is accepted

• ❌ Any non-developer, sensitive, or unrelated data is automatically rejected

• 🧠 A dedicated validation layer understands the context of incoming data and enforces these rules before storage

This ensures that Renard remains safe, privacy-aware, and developer-focused by design.