SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsx-hashtag-research

x-hashtag-research

Writing & research

View repository

Updated 10-15 16:47

About this skill

X Hashtag Research

A workspace skill that collects public posts starting from the FxTwitter API and organizes primary-source URLs, related GitHub repos, images, and discussion points under research.

When to Use

  • When you want to follow announcements from the day of an event via X hashtags
  • #MSBuild や #MicrosoftBuild When you want to sweep through an event hashtag like over the last few hours
  • When you want to gather primary-source URLs and repo leads first, right after a keynote
  • When you want to converge from noisy live posts toward Microsoft Learn / blog / GitHub / session repo
  • When you also want to keep image-attached posts as research assets

Scope

  • Make this skill X-only
  • Bluesky, LinkedIn, YouTube comments, etc. are out of scope
  • If you want to handle those, split them out into a separate skill

Default Profile

These are the default candidates when nothing is explicitly specified.

  • time window: 直近 4 時間
  • target posts: 500
  • raw artifact: tmp/
  • research note: research/YYYYMMDD-<topic>-from-x.md
  • curated images: research/assets/YYYYMMDD-<topic>-from-x/
  • bulk images: research/images/YYYYMMDD-<topic>-from-x/

Always Confirm Before Running

There are defaults, but confirm these 4 points every time.

  1. Target hashtag
  2. Time window
  3. Target post count
  4. Destination note name

If the user omits them, propose the default values and get confirmation. Do not run with 4時間 / 500件 on your own.

Collection Paths

  • For collecting public posts, limit to the agreed time window and post count, and use GET https://api.fxtwitter.com/2/search?q=<URL-encoded-query>&feed=latest&count=100 by default. Pass the next page to cursor.bottom を cursor .
  • FxTwitter is not X's official API. Do not pass cookies, tokens, or account information, and limit use to one-off collection of public posts.
  • Not just the HTTP status, but the JSON code Check , and adopt only200 の results . For400、404、500 , do not retry repeatedly; go back to the X screen or the primary source.
  • Do not use it for private, deleted, or access-restricted posts, or for continuous monitoring.
  • If the API fails, or if you need to check the on-screen context of a post, open X live search with a browser-based tool, get the post URL, time, body text, and image URL, and scroll down.
  • Even when structured extraction is weak, keep the order fixed: save raw posts, do rough classification by visible domain / card title, then dig deeper into high-signal primary sources.
  • Do not insist on full automation; prioritize official accounts, live blog leads, and representative images.

The core of this skill is not browser manipulation tips, but the order of converging SNS noise into primary sources.

Workflow

  1. Fix the collection targets Decide the target hashtag, time window, post count goal, main objective, and research save destination.

  2. Collect efficiently from the API From the API's results , extract structured data using the status URL as the primary key, and deduplicate. Only supplement from the screen if the API fails.

  3. Save raw artifacts first Keep tmp/<slug>-posts-raw.json and intermediate JSON per batch.

  4. Do local classification first Create theme, domain, repo, and image candidates from rawText, tweetText, display text, and the X card's visible domain.

  5. Dig deep only into high-signal links Do not resolve every t.co; follow only high-frequency links, official accounts, high-signal posts with images, and posts where half of the repo name is readable.

  6. Write research notes source-centric Make the primary source that the post converged on the canonical content, not impressions of the post.

  7. Save images in 2 layers Aim for roughly 5–10 images in the curated set and about 20–30 in bulk.

  8. Finish with reproducible artifacts Have everything ready: research notes, raw JSON, key primary-source URLs, image save locations, and if necessary, additions to the manifest.

Branching Rules

  • Public, one-time, capped collection: use the FxTwitter API search, andcode: 200 keep only the results of in the raw artifact
  • If the API fails or the post's on-screen context is needed: prioritize official accounts, live blogs, official blogs, and posts with a path to a repo
  • If resolving shortened URLs is heavy: do not resolve all of them. Resolve only high-frequency, high-signal ones
  • If there are too many images: secure curated first, and cap bulk

Quality Gates

  • Both the collection count and the time window are satisfied
  • The main themes are organized as a body of public information
  • If a repo name is cut off, it is corrected using the body text, card title, or official blog
  • The image save locations are organized under research
  • No assertions are made that alter the original public posts
  • When FxTwitter is used, the source is not the API URL but each status.url 's original X post URL

Efficiency Rules

  • tmp/*.json save intermediate results to avoid re-fetching
  • Do local classification first, then external fetches after
  • If there is a cross-cutting article on the official blog, use it as a hub and descend to individual articles
  • Do not hold onto too many browser-specific tips. If necessary, browser-max-automation use

Example Prompts

  • /x-hashtag-research #MSBuild と #MicrosoftBuild を直近4時間で500件以上集めて、一次情報と repo と画像を research に保存して
  • /x-hashtag-research #Build2026 で GitHub Copilot app と Foundry まわりだけ拾って
  • /x-hashtag-research #Ignite と #MicrosoftIgnite を2時間で300件、画像は代表8枚だけでいい

Related Customizations To Create Next

  • Prompt to insert hashtag research results into keynote research
  • Skill to select only article-suitable ones from saved images
  • Prompt to secondarily organize only official blog / Learn / GitHub repo