get_my_profileLABSCO SUMMARY
An agent can work your Twitter account, not just read it.
The timeline, the mentions and the search come back into the conversation, and the reply goes out from the same place — with the option of having the draft written first, if you add the paid key.
What it is
A Twitter/X client for agents, installed from PyPI. The read-and-post calls go straight at the Twitter API with your own OAuth tokens; the drafting and analytics calls route through ShipPost's service and need its key as well.
What you get
- Your own profile, your home timeline, and the tweets mentioning you
- Any user's profile, any tweet by id, and recent tweets by search
- Tweets posted, replies sent, tweets liked, users followed
- Your timeline scanned for the posts worth replying to (ShipPost key)
- Replies, single tweets and full threads drafted in your voice (ShipPost key)
- An account analysed, and engagement analytics reported back (ShipPost key)
Requirements
pip install shippost-mcp, plus a Twitter OAuth access token in the client config; a refresh token and your app's client id keep it renewing itself. The drafting and analytics calls additionally need a ShipPost API key.
Setup effort
One command plus a key — pip install shippost-mcp, then supply credentials
USE CASES
01
Catching up on what you missed
Read the home timeline and the mentions together, so what needs an answer and what is just noise separate in one pass.
02
Answering the mentions that deserve it
Pull the mentions, decide which are worth a response, and reply to each without leaving the conversation.
03
Posting the thing you just thought of
Say it once and have it go out, instead of switching apps and losing the sentence on the way.
04
Finding a conversation worth joining
Search on the topic, or have the timeline scanned for the best posts to reply to, then send the reply.
05
Getting a thread written before you commit to it
Have the thread drafted in your voice, read it, and post it once it sounds like you.
06
Sizing up an account before you engage
Read the public profile, then get the account analysed, so an outreach decision is made on something more than follower count.
17 tools
Listed in project docs. Your client may show a different list.
get_timelineget_mentionsget_user_profileget_tweetsearch_tweetspost_tweetreply_to_tweetlike_tweetretweetfollow_userfind_opportunitiesdraft_replydraft_tweetdraft_threadanalyze_accountget_performanceBEFORE YOU INSTALL
What to know first
Tools you get
17 tools named in the project's documentation.
Requires
- shippost-mcp on PATH
- TWITTER_ACCESS_TOKEN in your environment
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
pip install shippost-mcpMCP Clients
Claude Code
CLAUDE CODE · JSON
{
"mcpServers": {
"twitter": {
"command": "shippost-mcp",
"env": {
"TWITTER_ACCESS_TOKEN": "<YOUR_TWITTER_ACCESS_TOKEN>",
"TWITTER_REFRESH_TOKEN": "<YOUR_TWITTER_REFRESH_TOKEN>",
"TWITTER_CLIENT_ID": "<YOUR_TWITTER_CLIENT_ID>"
}
}
}
}
