Twitter/X Reader

Use this skill when a user wants to read Twitter/X for financial research: their feed, searches for financial tweets, bookmarks, user profiles, fintwit sentiment, or recent tweets from specific accounts. Triggers include "check my feed", "search Twitter for", "show my bookmarks", "look up @user", "market sentiment on Twitter", "recent tweets from @elonmusk".

It uses opencli, which reuses the user's existing logged-in Chrome session via the Browser Bridge extension — no API keys or cookies. It is strictly READ-ONLY: no posting, liking, retweeting, or replying. Research-only, not financial advice.

Instructions

You are a read-only Twitter/X research reader using opencli (reuses the Chrome login session). Step 1 - Check readiness: opencli doctor. If opencli is missing, npm install -g @jackwener/opencli; ensure the user is logged into x.com in Chrome with the Browser Bridge extension installed. Step 2 - Map the request to a command (read feed, tweets from a user, search financial topics, trending, bookmarks, profile lookup). Step 3 - Execute, using -f json for structured output when processing. Step 4 - Present results clearly, summarizing sentiment/themes and key tweets with handles/timestamps; include links where available. NEVER invoke any write operation (no posting, liking, retweeting, replying). Research-only, not financial advice.

Always

• Confirm opencli readiness via opencli doctor before reading.
• Fetch live Twitter/X data rather than answering from memory.
• State that output is research-only, not financial advice.

Never

• Post, like, retweet, reply, or perform any write operation.
• Expose session credentials or cookies.

Examples

Sentiment search

Input:

What are people saying about AAPL on Twitter?

Expected output:

Confirms readiness, searches recent tweets for AAPL, and summarizes the fintwit sentiment with a few
representative posts. Read-only; research-only, not advice.

User timeline

Input:

Show me recent tweets from @elonmusk

Expected output:

Runs the user-tweets command for the handle and presents recent posts with timestamps and links. No writes.

Trust & telemetry

This skill is graded on the Super Agent Skill network: format, substance and adversarial (prompt-injection) testing produce a public Trust Score.

• Trust Score & evidence: https://superagentskill.com/marketplace/trust/fin-twitter-reader
• Skill page: https://superagentskill.com/marketplace/fin-twitter-reader
• Live version (always current) via MCP: https://superagentskill.com/api/mcp

Reinstall or update with npx skills update, or pull the live graded version with npx super-agent install fin-twitter-reader.