Sync and store locally all of your X/Twitter bookmarks. Search, classify, and make them available to Claude Code, Codex, or any agent with shell access.

Free and open source. Designed for Mac.

npm install -g fieldtheory

Requires Node.js 20+. A Chrome-family browser or Firefox is recommended for session sync; OAuth is available for all platforms.

# 1. Sync your bookmarks (needs a supported browser logged into X)
ft sync

# 2. Search them
ft search "distributed systems"

# 3. Explore
ft viz
ft categories
ft stats

On first run, ft sync extracts your X session from your browser and downloads your bookmarks into ~/.fieldtheory/bookmarks/.

ft library open targets the packaged Field Theory app by bundle id (com.fieldtheory.app) instead of trusting the system-wide fieldtheory:// handler. That avoids accidentally opening a generic Electron development app when another checkout registered the same URL scheme.

For local Field Theory app development, point the CLI at the dev checkout:

export FT_APP_DEV_DIR=/Users/you/dev/fieldtheory/mac-app
ft library open notes/example.md

Packaged variants can override the bundle id with FT_APP_BUNDLE_ID. Advanced development launchers can set FT_APP_OPEN_COMMAND to an executable that receives the deep-link URL as its first argument.

Install the /fieldtheory skill so your agent automatically searches your bookmarks when relevant:

ft skill install     # Auto-detects Claude Code and Codex

Then ask your agent:

"What have I bookmarked about cancer research in the last three years and how has it progressed?"

"I bookmarked a number of new open source AI memory tools. Pick the best one and figure out how to incorporate it in this repo."

"Your goal is to look at AI agent bookmarks and come up with a roadmap plotted in the grid of what I should do next across the Field Theory CLI and Mac app projects."

"Every day please sync any new X bookmarks using the Field Theory CLI."

Works with Claude Code, Codex, or any agent with shell access.

Sync with cron:

# Sync every morning at 7am
0 7 * * * ft sync

# Sync and classify every morning
0 7 * * * ft sync --classify

Run Possible every night on macOS with LaunchAgent:

ft seeds search "agents" --days 90 --limit 8 --frame leverage-specificity --create
ft repos add ~/dev/fieldtheory
ft repos add ~/dev/fieldtheory-cli

ft possible nightly install --time 02:00 --defaults --model opus --effort medium --nodes 5
ft possible nightly show

Nightly schedules are stored under ~/.fieldtheory/ideas/nightly/. Each tick starts a normal background job under ~/.fieldtheory/ideas/jobs/, using your local logged-in CLI sessions and the current PATH captured in the LaunchAgent plist.

ft respects standard proxy environment variables for network requests: HTTPS_PROXY, HTTP_PROXY, ALL_PROXY, and NO_PROXY.

Data is stored locally under ~/.fieldtheory/:

~/.fieldtheory/bookmarks/
  bookmarks.jsonl         # raw bookmark cache (one per line)
  bookmarks.db            # SQLite FTS5 search index
  bookmarks-meta.json     # sync metadata
  oauth-token.json        # OAuth token (if using API mode, chmod 600)

~/.fieldtheory/library/
  index.md                # markdown knowledge base (ft wiki / ft md)

~/.fieldtheory/commands/
  *.md                    # portable commands used by Field Theory and agents

~/.fieldtheory/ideas/
  seeds/runs/nodes/       # Possible seeds, runs, and node prompt artifacts
  batches/jobs/nightly/   # Multi-repo batches, background jobs, and schedules

Override locations with FT_DATA_DIR, FT_LIBRARY_DIR, and FT_COMMANDS_DIR:

export FT_DATA_DIR=/path/to/custom/dir
export FT_LIBRARY_DIR=/path/to/custom/library
export FT_COMMANDS_DIR=/path/to/custom/commands

To remove bookmark and Library data: rm -rf ~/.fieldtheory/bookmarks ~/.fieldtheory/library

Use ft classify for LLM-powered classification that catches what regex misses.

In PowerShell, use fieldtheory or ft.cmd instead of ft because ft is already a built-in alias for Format-Table.

If browser session sync cannot find the right profile, pass the browser and profile explicitly:

fieldtheory sync --browser chrome --chrome-profile-directory "Default"
fieldtheory sync --browser edge --chrome-profile-directory "Default"

For Firefox, if profile detection misses the profile, pass the profile directory explicitly with --firefox-profile-dir.

If cookie extraction still fails, close the browser completely and retry. As a last resort, pass cookies manually:

fieldtheory sync --cookies <ct0> <auth_token>

Treat ct0 and auth_token like passwords. Do not paste them into logs, issues, or chat.

Session sync extracts cookies from your browser's local database. Use ft sync --browser <name> to pick a browser. On Windows, Firefox requires Node.js 22.5+ or sqlite3 on PATH. For unsupported browsers or platforms, use ft auth + ft sync --api.

Your data stays local. No telemetry, no analytics, nothing phoned home. The CLI only makes network requests to X's API during sync.

Chrome session sync reads cookies from Chrome's local database, uses them for the sync request, and discards them. Cookies are never stored separately.

OAuth tokens are stored with chmod 600 (owner-only). Treat ~/.fieldtheory/bookmarks/oauth-token.json like a password.

The default sync uses X's internal GraphQL API, the same API that x.com uses in your browser. For the official v2 API, use ft auth + ft sync --api.

MIT — fieldtheory.dev/cli