1. Use mcp__claude-in-chrome__navigate to go to https://x.com
2. Wait for page to load (use mcp__claude-in-chrome__computer with action: "wait")
3. Take screenshot to verify page loaded

1. Navigate to https://x.com/search
2. Use mcp__claude-in-chrome__find to locate the search input
3. Use mcp__claude-in-chrome__form_input to enter search query
4. Press Enter to submit search
5. Wait for results to load

1. Navigate to https://x.com/{username}
2. Wait for profile to load
3. Verify profile loaded by checking for user's name/handle

1. Use mcp__claude-in-chrome__computer with action: "scroll" and scroll_direction: "down"
2. Take screenshot after scrolling
3. Repeat as needed to gather more content

1. Use mcp__claude-in-chrome__find to locate the tweet
2. Click on the tweet to open detail view
3. Wait for replies and engagement data to load

1. Use mcp__claude-in-chrome__read_page to get page accessibility tree
2. Look for article elements containing tweet content
3. Extract text, author, timestamp, and engagement metrics

1. Navigate to profile page
2. Use mcp__claude-in-chrome__read_page
3. Extract: name, handle, bio, follower/following counts, pinned tweet