browsing-x

name	browsing-x
description	Use when the user wants to interact with X (formerly Twitter) — search for posts, extract a specific post's data (content + engagement + author), or extract a user's profile (bio, followers, etc.). Works on x.com and twitter.com. Requires a logged-in session (auth_token cookie) and a real browser — X is a JS-rendered SPA.

X (Twitter) — Browsing Skill

Use this index to choose the X action that matches the user request, then open the linked reference file for the complete navigation, requirements, code, and return shape.

Action Index

• search — Search X for a query and extract visible post results. Full spec: references/search.md.
• post-data — Extract a specific post's content, author details, engagement metrics, timestamp, and media. Full spec: references/post-data.md.
• profile-data — Extract an X user profile, including bio, follower counts, avatar, website, and account metadata. Full spec: references/profile-data.md.

Benchmarks

Benchmarks compare the maintained skill action against a no-skill browser agent that inspects the live page DOM and derives selectors at runtime. Full notes live in BENCHMARKS.md.

Action	With Skill	Without Skill	Notes
search	1,699 / ~6.4s	3,907 / ~15.8s	Skill valid; no-skill malformed partial result.
post-data	1,987 / ~4.3s	14,333 / ~35.1s	Skill complete; no-skill partial result.
profile-data	1,865 / ~12.3s	324 / failed	Skill complete; no-skill one-shot eval failed.