Summary

Field Theory CLI is a free, open-source command-line tool designed to sync, store, and manage X/Twitter bookmarks locally. It empowers users, particularly on macOS, to search, classify, and integrate their bookmarked content with AI agents, offering a robust alternative to X's native bookmarking.

How It Works

The core approach leverages Chrome session extraction on macOS for initial bookmark synchronization without requiring API keys, storing all data locally in ~/.ft-bookmarks/. This includes raw bookmarks, an SQLite FTS5 search index, and metadata. For broader compatibility or API-based sync, it supports OAuth authentication. Bookmarks can be classified using LLMs or regex, and a terminal dashboard provides interactive exploration.

Quick Start & Requirements

• Install: npm install -g fieldtheory
• Prerequisites: Node.js 20+, Google Chrome. macOS Keychain is used for Chrome session extraction.
• Run: Execute ft sync to download bookmarks.
• Data Location: Stored locally at ~/.ft-bookmarks/.
• Docs: fieldtheory.dev/cli (implied).

Highlighted Details

• API-Free Sync: Initial sync via Chrome session extraction (macOS) avoids API key management.
• LLM Classification: Advanced categorization using LLMs (ft sync --classify) complements regex-based methods.
• Powerful Search: Full-text search powered by BM25 ranking (ft search).
• Agent Integration: Seamlessly integrates with AI agents like Claude Code and Codex for natural language querying of bookmarks.
• Interactive Dashboard: Visualize categories, domains, and stats within the terminal (ft viz).
• Cross-Platform API Sync: Optional OAuth sync (ft sync --api) enables use on Linux and Windows.
• Media Fetching: Supports downloading static images associated with bookmarks (ft fetch-media).

Maintenance & Community

No specific details regarding maintainers, sponsorships, or community channels (like Discord/Slack) were found in the provided README.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration into closed-source projects.

Limitations & Caveats

The primary bookmark sync method relies on Chrome session extraction, which is currently limited to macOS. Users on Linux and Windows must utilize the ft auth and ft sync --api commands for synchronization. Media fetching is restricted to static images.