Skip to content

COMMANDS.md

Latest commit

 

History

History
270 lines (194 loc) · 5.96 KB

COMMANDS.md

File metadata and controls

270 lines (194 loc) · 5.96 KB

BNBot CLI Command Reference

Overview

bnbot <command> [subcommand] [args] [options]

BNBot CLI communicates with the BNBot Chrome Extension via WebSocket to control Twitter/X. Public data scrapers work standalone without the extension.

Setup & Authentication

Command Description
bnbot setup Install CLI globally + Claude Code skill
bnbot login Login (auto-detects clawmoney API key)
bnbot login --email you@example.com Login via email verification
bnbot serve Start WebSocket server
bnbot serve --port 18901 Start on custom port
bnbot status Check extension connection

X Platform Commands

All X commands require the Chrome extension to be connected (bnbot serve running + extension open on Twitter).

Post & Compose

# Post a tweet
bnbot x post "Hello from BNBot!"

# Post with media (local file or URL)
bnbot x post "Check this out" -m ./photo.png
bnbot x post "Look" -m https://example.com/image.jpg

# Multiple media (auto-splits into thread if >4)
bnbot x post "Gallery" -m img1.png -m img2.png -m img3.png

# Draft mode (fill composer without posting)
bnbot x post "Draft tweet" -d

# Post a thread (JSON array)
bnbot x thread '[{"text":"First tweet"},{"text":"Second tweet"}]'

# Close composer
bnbot x close
bnbot x close --save  # Save as draft

Engage

# Reply to a tweet
bnbot x reply https://x.com/user/status/123 "Great post!"
bnbot x reply https://x.com/user/status/123 "With media" -m photo.jpg

# Quote tweet
bnbot x quote https://x.com/user/status/123 "My thoughts on this"

# Like / Unlike
bnbot x like https://x.com/user/status/123
bnbot x unlike https://x.com/user/status/123

# Retweet / Unretweet
bnbot x retweet https://x.com/user/status/123
bnbot x unretweet https://x.com/user/status/123

# Follow / Unfollow
bnbot x follow elonmusk
bnbot x unfollow elonmusk

# Delete a tweet
bnbot x delete https://x.com/user/status/123

# Bookmark / Unbookmark
bnbot x bookmark https://x.com/user/status/123
bnbot x unbookmark https://x.com/user/status/123

Scrape

# Scrape home timeline
bnbot x scrape timeline
bnbot x scrape timeline -l 50

# Scrape bookmarks
bnbot x scrape bookmarks
bnbot x scrape bookmarks -l 10

# Search and scrape
bnbot x scrape search "AI agents"
bnbot x scrape search "AI agents" -t latest -l 50
bnbot x scrape search "AI" --from elonmusk --since 2025-01-01 --until 2025-12-31
bnbot x scrape search "AI" --lang en --minLikes 100 --has images

# Scrape user tweets
bnbot x scrape user-tweets elonmusk
bnbot x scrape user-tweets elonmusk -l 50

# Scrape user profile
bnbot x scrape user-profile elonmusk

# Scrape a thread
bnbot x scrape thread https://x.com/user/status/123

Search options:

Option Description Default
-t, --tab <tab> Search tab: top, latest, people, media top
-l, --limit <n> Max results 20
--from <username> Filter by author
--since <date> Start date (YYYY-MM-DD)
--until <date> End date (YYYY-MM-DD)
--lang <code> Language filter (en, zh, etc.)
--minLikes <n> Minimum likes
--minRetweets <n> Minimum retweets
--has <type> Media filter: images, videos, links

Analytics

bnbot x analytics

Navigate

# Navigate to a URL
bnbot x navigate https://x.com/elonmusk/status/123

# Navigate to search
bnbot x navigate search "AI agents"

# Navigate to bookmarks
bnbot x navigate bookmarks

# Navigate to notifications
bnbot x navigate notifications

Public Data Scrapers

These commands fetch data directly from public APIs. No Chrome extension or WebSocket server needed.

Hacker News

bnbot hackernews search "AI agents"
bnbot hackernews search "LLM" -l 10
bnbot hackernews search "startup" --sort date

Stack Overflow

bnbot stackoverflow search "react hooks"
bnbot stackoverflow search "typescript generics" -l 5

Wikipedia

bnbot wikipedia search "artificial intelligence"
bnbot wikipedia search "ChatGPT" --lang zh

Apple Podcasts

bnbot apple-podcasts search "tech news"

Substack

bnbot substack search "AI newsletter"

V2EX

bnbot v2ex hot

Bloomberg

bnbot bloomberg news
bnbot bloomberg news -l 10

BBC

bnbot bbc news
bnbot bbc news -l 10

Sina Finance

bnbot sinafinance news
bnbot sinafinance news --type 1 -l 10

Sina Blog

bnbot sinablog search "AI"

Xiaoyuzhou FM

bnbot xiaoyuzhou podcast 6021f949ee2c6723b18a13b4

Content Fetching (via Extension)

These commands require the Chrome extension.

# WeChat article
bnbot weixin article https://mp.weixin.qq.com/s/xxx

# TikTok video
bnbot tiktok fetch https://www.tiktok.com/@user/video/xxx

# Xiaohongshu note
bnbot xiaohongshu fetch https://www.xiaohongshu.com/explore/xxx

Backward Compatibility

Legacy kebab-case commands from previous versions still work:

# These are equivalent:
bnbot post-tweet --text "Hello"    # legacy
bnbot x post "Hello"               # new

bnbot scrape-timeline --limit 10   # legacy
bnbot x scrape timeline -l 10      # new

bnbot like-tweet --tweetUrl <url>  # legacy
bnbot x like <url>                 # new

bnbot search-hackernews --query "AI"  # legacy
bnbot hackernews search "AI"          # new

Architecture

bnbot serve                    ← Start WebSocket server on ws://localhost:18900
    ↑
bnbot x post "Hello"           ← CLI client connects, sends action, exits
    ↓
Chrome Extension (BNBot)       ← Executes action in browser, returns result
    ↓
Twitter / X                    ← DOM manipulation / API calls
  • bnbot serve starts a persistent WebSocket server
  • CLI commands (bnbot x ...) connect as transient clients
  • The Chrome extension connects to the server and executes actions
  • Public scrapers (bnbot hackernews ...) bypass WebSocket entirely