llm-lean-log

Thellm-lean-logproject offers a smart way to keep track of your conversations with AI. It's designed to save you money by using fewer words, or tokens, when recording these chats. This makes it great for building your own reference library, documenting projects, or keeping your chat history organized across different computers.

At its heart,llm-lean-loguses a special format that is like a simple spreadsheet, called CSV. This format is very efficient for AI to understand and is more cost-effective than other formats like Markdown or JSON. It focuses on how different parts of a conversation relate to each other, like cause and effect.

Benefits

• Saves Money:By using a token-efficient format, it lowers the cost of logging AI chats.
• Organized History:It helps you build a personal knowledge base and document your work.
• Easy to Use:Command-line tools make it simple to add, view, search, and filter your chat logs.
• Detailed Records:Logs can include information like the problem, solution, actions taken, and files changed.
• Code Integration:It can automatically save code changes made during a chat session.
• Visualizations:Chat logs can be viewed in a user-friendly web format with code highlighting.

Use Cases

llm-lean-logis useful for anyone working with AI agents who needs to keep detailed records. This includes developers documenting their coding process, researchers tracking AI experiments, or individuals building a personal knowledge base from AI interactions. The command-line tools allow users to easily add new log entries, specifying details about tasks, problems, and solutions. It can also be integrated into AI agent workflows to automatically save important information as the agent works.

Vibes

The project is open-source, licensed under MIT, and aims to provide a practical solution for managing AI chat logs efficiently. It includes tools for both machine readability (CSV format for AI) and human readability (visualizer and table output).

Additional Information

The project provides several command-line tools.llm-lean-log-cli(orl-log) is used for managing the logs.l-log-visoffers a web-based visualizer to see the logs in a clear, human-friendly way. There is also a server for MCP memory functionality. The format includes specific fields like ID, name, tags, problem, solution, actions, files affected, and the technology stack used, along with timestamps and the AI model involved.