Twitter X Hub

What it does

When your agent uses it

• Mentions scraping Twitter data
• Twitter timeline
• Request Twitter with cookies
• Twitter GraphQL

Example prompts

• “scraping Twitter data,”
• “get X tweets,”
• “Twitter timeline,”
• “/twitter-x-hub”

Requirements

• Python 3
• A credential in TWITTER_AUTH_TOKEN
• A credential in COOKIE_AUTH_TOKEN

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

1. 1Use browser_use navigate to open https://x.com and confirm you are logged in.
2. 2Use browser_use get_cookies to retrieve all cookies.
3. 3Load the file, then use the cookies

What it can do on your machine

Read from SKILL.md and the folder at commit ae8c5db. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships 5 files in scripts/ (Python), which the agent can run.

  Shell commands in SKILL.md call:

  • uv
  • pip

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Hosts in commands or code, which the agent is likely to contact:

  • x.com

  Also links to:

  • github.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • TWITTER_AUTH_TOKEN
  • COOKIE_AUTH_TOKEN

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Twitter X Hub loads about 2.9k tokens when it runs. Until then it costs about 249 tokens; SKILL.md has 803 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~249

When it runs · the whole SKILL.md, loaded when a task matches

~2.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from OpenMinis/MinisSkills at commit ae8c5db, republished under its MIT licence (© OpenMinis). 803 words, ~2,917 tokens.

name

twitter-x-hub

description

Skill for reading and writing Twitter/X data with Python + UV, with zero third-party dependencies (pure standard library), by directly passing the auth_token + ct0 cookies for authentication. In the Minis environment, cookies can be retrieved automatically with the browser_use tool by navigating to x.com and then using the get_cookies action, with no manual copying required. Supports scraping the home timeline, following list, bookmarks (including bookmark folders), search, user profiles, user tweets, likes, tweet details (single tweet or with replies), List timelines, followers/following lists, and write operations such as posting, deleting, liking, retweeting, and bookmarking. This skill must be triggered when the user mentions "scraping Twitter data," "get X tweets," "Twitter timeline," "X bookmarks," "search tweets," "twitter-x-hub," "request Twitter with cookies," "Twitter GraphQL," or any scenario that requires programmatic reading or writing of Twitter/X data.

/var/minis/skills/twitter-x-hub/
├── SKILL.md
├── pyproject.toml              # UV project configuration (zero third-party dependencies)
└── scripts/
    ├── __init__.py
    ├── models.py               # Data models (Tweet, Author, Metrics, UserProfile, BookmarkFolder)
    ├── parser.py               # GraphQL response parsing (split from client.py, synced from upstream v0.8.6)
    ├── client.py               # GraphQL client (core logic)
    └── cli.py                  # Command-line entry point (argparse)

. /var/minis/offloads/env_cookies_xxx.sh
export TWITTER_AUTH_TOKEN="$COOKIE_AUTH_TOKEN"
export TWITTER_CT0="$COOKIE_CT0"

# Confirm UV is available
which uv || pip install uv

# Enter the skill directory
cd /var/minis/skills/twitter-x-hub

# Scrape the home For-You timeline (20 items by default)
uv run python -m scripts.cli feed

# Scrape the Following timeline, 30 items, JSON output
uv run python -m scripts.cli feed --type following --max 30 --json

# Search tweets (Top/Latest/Photos/Videos)
uv run python -m scripts.cli search "Claude Code" --tab Latest --max 20

# Bookmarks
uv run python -m scripts.cli bookmarks --max 50

# Bookmark folder list (new)
uv run python -m scripts.cli bookmark-folders

# User profile
uv run python -m scripts.cli user elonmusk

# User tweets
uv run python -m scripts.cli user-posts elonmusk --max 20

# User likes
uv run python -m scripts.cli user-likes elonmusk --max 20

# Tweet details (including reply thread)
uv run python -m scripts.cli tweet 1234567890

# Quickly retrieve a single tweet (new, faster than the tweet command)
uv run python -m scripts.cli tweet-by-id 1234567890

# List timeline
uv run python -m scripts.cli list 1539453138322673664

# Followers / following lists (first use the user command to get user_id)
uv run python -m scripts.cli followers <user_id> --max 50
uv run python -m scripts.cli following <user_id> --max 50

# Post a tweet / reply
uv run python -m scripts.cli post "Hello from twitter-x-hub!"
uv run python -m scripts.cli post "reply text" --reply-to 1234567890

# Like / retweet / bookmark
uv run python -m scripts.cli like 1234567890
uv run python -m scripts.cli retweet 1234567890
uv run python -m scripts.cli bookmark 1234567890

export TWITTER_AUTH_TOKEN="xxxx"
export TWITTER_CT0="yyyy"

uv run python -m scripts.cli feed --max 30 --json

import os, json, dataclasses
from scripts.client import TwitterClient

client = TwitterClient(
    auth_token=os.environ["TWITTER_AUTH_TOKEN"],
    ct0=os.environ["TWITTER_CT0"],
)

# Scrape the home timeline
tweets = client.fetch_home_timeline(count=20)
for t in tweets:
    print(f"@{t.author.screen_name}: {t.text[:80]}")
    print(f"  ❤️ {t.metrics.likes}  🔁 {t.metrics.retweets}  👁 {t.metrics.views}  🔖 {t.metrics.bookmarks}")

# Search (Latest tab)
results = client.fetch_search("AI agent", count=10, product="Latest")

# Single tweet (quick, no replies)
tweet = client.fetch_tweet_by_id("1234567890")

# Bookmark folders
folders = client.fetch_bookmark_folders()

# User profile
user = client.fetch_user("elonmusk")
print(user.id, user.followers_count)

# JSON serialization
data = [dataclasses.asdict(t) for t in tweets]
print(json.dumps(data, ensure_ascii=False, indent=2))

1. In-memory cache (fastest)
2. Hard-coded FALLBACK_QUERY_IDS (constant fallback)
   → If 404, the queryId has expired; proceed to the next level
3. Fetch the latest queryId from github.com/fa0311/twitter-openapi
   → If it is still unavailable, scan the x.com JS Bundle and extract it with a regular expression

© OpenMinis, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 7 other files (scripts) in twitter-x-hub of OpenMinis/MinisSkills.

• SKILL.md
• pyproject.toml
• scripts/__init__.py
• scripts/cli.py
• scripts/client.py
• scripts/models.py
• scripts/parser.py
• uv.lock

Open the folder on GitHubat commit ae8c5db