afar1/fieldtheory-cli

Field Theory CLI for bookmarks, Library, commands, and agent workflows

What it solves

Field Theory CLI solves the problem of fragmented digital bookmarks by syncing X (Twitter) bookmarks to a local, searchable, and LLM-enhanced knowledge base. It allows users to move their curated content out of a proprietary platform and into a local environment where it can be classified, exported to Markdown, and used as context for AI agents.

How it works

The tool extracts session cookies from supported browsers or uses OAuth to sync bookmarks from X into a local SQLite database and JSONL cache. It then provides a suite of commands to search, filter, and visualize this data. Users can use LLMs to automatically classify bookmarks into categories (like research, tools, or techniques) and export them into a "Karpathy-style" interlinked Markdown wiki. Additionally, it can install "skills" that allow AI agents (like Claude Code or Codex) to query the local bookmark library via the shell.

Who it’s for

It is designed for power users, researchers, and developers (primarily on macOS) who use X for curation and want a local, agent-accessible knowledge base of their bookmarks.

Highlights

  • Local-First Storage: All data is stored in ~/.fieldtheory, ensuring privacy and no telemetry.
  • LLM Classification: Automatically categorizes bookmarks using LLM engines to organize content beyond simple tags.
  • Agent Integration: Provides a /fieldtheory skill for AI agents to search and utilize bookmarks as grounded context.
  • Knowledge Base Export: Converts bookmarks into an interlinked Markdown wiki for long-term knowledge management.
  • Browser Session Sync: Syncs bookmarks without requiring an official API key by extracting session data from Chrome-family or Firefox browsers.

Related

  • Project
    vercel-labs/lat.md

    lat.md is an npm‑distributed CLI that creates a markdown‑based knowledge graph for a codebase. It lets you write linked documentation in `lat.md/`, annotate source files with `@lat` comments, and validates the graph. Agents and developers can search it semantically (offline MiniLM or remote OpenAI/Vercel embeddings) and use commands like `lat check`, `lat locate`, and `lat expand` to keep design decisions, constraints, and test specs in sync with the code.

  • Project
    ussumant/llm-wiki-compiler

    A Claude Code and Codex plugin that compiles scattered markdown files and codebases into a synthesized, topic-based wiki to reduce AI agent context costs.

  • Project
    sdyckjq-lab/llm-wiki-skill

    A system for building a structured, interconnected knowledge base from fragmented data, featuring an interactive knowledge graph and integration with AI CLI agents.

  • Project
    langchain-ai/openwiki

    A CLI tool that uses AI agents to create and maintain a self-updating Markdown wiki for codebases or personal knowledge, featuring grounded claims to track facts back to source evidence.

  • Project
    nashsu/llm_wiki

    A desktop application that automatically transforms documents into a structured, interlinked personal wiki using LLMs, featuring a knowledge graph and automated deep research.