Mmcp.market

x-tweet-by-conversation skill

by browser-act·browser-act/skills·6.0k stars·MIT

Collects every tweet in an X (Twitter) conversation thread given a conversation id (root tweet id) — the focal tweet plus all replies, sub-replies, and quote chains — and returns normalized per-tweet data with text, author, engagement counts, media, hashtags, mentions, in_reply_to mapping, and cursor for pagination. Use when user mentions Twitter conversation, X conversation thread, conversation_id, thread scraper, scrape Twitter replies, all replies to a tweet, replies under a tweet, sub-replies, nested replies, thread harvester, get replies of a tweet, scrape comments on Twitter, scrape comments on X, full thread extraction, conversation export, conversation tree, reply chain, thread dump, X tweet thread, twitter thread scrape, comment scraping twitter, comment scraping x, focal tweet plus context, root tweet plus replies. Also applies to sentiment analysis on a single viral tweet, controversy mapping, harvesting community Q&A threads, capturing AMA threads, recovering long-running discussions, and any paginated bulk reply collection driven by a conversation id.

A97/100content scan

Is the x-tweet-by-conversation skill safe?

Clean: nothing in its files matched our rules. We read 3 files in the folder on 2026-09-28.

  • lowSKILL.md:1

    The description is over 1,024 characters, the limit agents read.

    1080 characters

Install the x-tweet-by-conversation skill

A skill is a folder. Copy it into your agent's skills folder and the agent loads it when the task matches its description. 

git clone --depth 1 https://github.com/browser-act/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/solutions/social-listening/x-tweet-by-conversation ~/.claude/skills/x-tweet-by-conversation
available in every project

In the Claude apps, zip the folder and upload it from the Skills settings. The folder on GitHub

The instructions your agent would load

SKILL.md as published, without the frontmatter. Read it on GitHub

X — Tweets by Conversation

Conversation id (root tweet id) → normalized list of every tweet in the thread (focal tweet + replies + sub-replies), with author, engagement, media, cursor.

Language

All process output to user (progress updates, process notifications) follows the user's language.

Objective

Given an X conversation id (which equals the root tweet id), collect the focal tweet and every reply / sub-reply visible to the logged-in session, returning structured per-tweet data with pagination cursors.

Prerequisites

  • Active X session in the browser (left sidebar shows logged-in avatar / @handle).
  • Network capture is enabled in the browser-act session.

Pre-execution Checks

1. Tool Readiness

If browser-act has been confirmed available in the current session → skip this step.

Invoke browser-act via Skill tool to load usage. If installation or configuration issues arise, follow its guidance to resolve then retry.

2. Login Verification

If login status for X has been confirmed in the current session → skip this step.

Otherwise: open https://x.com and observe the left sidebar:

  • User avatar or @handle visible → logged in, continue
  • "Sign in" / "Log in" prompt visible → not logged in, inform the user and assist the login flow

User refuses or cannot log in → terminate execution.

Capability Components

This Skill's operational boundary = what the user can manually do in their browser. It only reads tweet data already shown to the user, never bypassing authentication. The browser's own JS signs the GraphQL request; the Skill triggers it via URL navigation and reads the response from network traffic. Python scripts under scripts/ only build URLs and parse responses — they do not call X directly. Run them through the bash tool.

Network Capture: full conversation thread

Step 1 — build the tweet-detail URL for the conversation root:

URL=$(python scripts/build-conversation-url.py '{conversation_id}' [--handle {handle}])

Parameters:

  • conversation_id (positional): the X conversation id, which equals the root tweet id (the same number you would see in https://x.com/i/status/).
  • --handle: the root tweet's author handle if known; if unknown, leave the default — x.com/i/status/ resolves to the same focal tweet because X canonicalises the URL.

Step 2 — navigate and capture the first page:

  1. network requests --clear
  2. navigate "$URL"
  3. wait stable --timeout 25000 (timeout is normal on X; proceed)
  4. network requests --type xhr,fetch --filter TweetDetail → take the latest entry's request_id
  5. network request → save full output to a file (e.g. tmp/x-conversation-page-1.txt)
  6. python scripts/parse-tweets.py --json-file tmp/x-conversation-page-1.txt --source tweetdetail → emits JSON {tweets, count, cursortop, cursorbottom}. The focal tweet is the first element; subsequent elements are replies (in display order). Each reply carries isreply: true, inreplytoid, inreplytouser, and conversation_id == , which makes the reply tree reconstructable downstream.

Endpoint characteristic: URL contains /i/api/graphql//TweetDetail. The query hash rotates; always filter by name.

Step 3 — paginate via scroll to load deeper replies and sub-replies:

  1. network requests --clear
  2. scroll down --amount 5000
  3. wait stable --timeout 10000
  4. network requests --type xhr,fetch --filter TweetDetail → newest entry's request_id
  5. network request → save to tmp/x-conversation-page-N.txt
  6. python scripts/parse-tweets.py --json-file tmp/x-conversation-page-N.txt --source tweet_detail

Repeat Step 3 until any termination condition is met:

  • Accumulated unique reply count reaches the user's target.
  • count == 0 on the current page.
  • cursor_bottom is unchanged across two consecutive pages.
  • The page shows a "Show more replies" button gated by visibility filters — state reveals it; the Agent may click to expand more, then resume scrolling.

Error handling:

  • If the focal tweet is deleted or protected, the response carries a TweetTombstone entry; parse-tweets.py filters tombstones, so the result will be count: 0 — report to the user and stop.
  • If no TweetDetail request appears after a scroll, wait 3 s and retry once. Persistent absence after two scrolls means the thread has been fully loaded.
  • If a captcha challenge appears (visible in state as an authorization prompt), pause and ask the user.

Output example:

{
  "tweets": [
    {
      "type": "tweet",
      "id": "2069990565530214798",
      "url": "https://x.com/Rothmus/status/2069990565530214798",
      "twitter_url": "https://twitter.com/Rothmus/status/2069990565530214798",
      "text": "\"We are going to have a multi-racial nation in Singapore. ...\"",
      "created_at": "Tue Jun 24 23:01:50 +0000 2026",
      "lang": "en",
      "source": "Twitter Web App",
      "retweet_count": 91,
      "reply_count": 84,
      "like_count": 567,
      "quote_count": 12,
      "bookmark_count": 50,
      "view_count": 64140,
      "is_reply": false,
      "is_retweet": false,
      "is_quote": false,
      "quote_id": null,
      "quote_url": null,
      "in_reply_to_id": null,
      "in_reply_to_user": null,
      "in_reply_to_user_id": null,
      "conversation_id": "2069990565530214798",
      "hashtags": [],
      "mentions": [],
      "urls": [],
      "media": [],
      "card": null,
      "place": null,
      "author": {
        "id": "987654321",
        "user_name": "Rothmus",
        "name": "Rothmus",
        "url": "https://x.com/Rothmus",
        "is_verified": false,
        "is_blue_verified": true,
        "verified_type": nu

Pagination

Network Capture Pagination: triggered by scroll down. X's page JS inserts the previous response's cursorbottom into the next TweetDetail request's variables.cursor. Some sub-reply branches require clicking an in-page "Show replies" expander (state to locate, click ) before the next scroll surfaces them. Termination: count == 0, cursorbottom does not advance across two consecutive pages, or user target reached.

Success Criteria

count >= 1 on the first page (the focal tweet must be present unless the conversation root is deleted) AND every tweet has non-null id, text, createdat, author.username, likecount, retweetcount, replycount, conversationid AND each reply tweet has isreply == true and inreplytoid pointing to a tweet inside the thread.

Known Limitations

  • Only replies visible to the logged-in session are returned; X enforces visibility filters (blocked accounts, soft-blocked replies, "Show additional replies, including those that may contain offensive content").
  • Deleted root tweets return a tombstone; the Skill terminates with count: 0.
  • Quote tweets that reference this conversation are NOT included — they live in a separate timeline; collect them via x-tweet-search-by-query with filter:quote conversation_id: if needed.
  • view_count is null for very new replies where X has not emitted views.count.
  • Sustained polling triggers per-session throttling; stay under ~150 timeline calls per 15-minute window per session.
  • The page may never reach network-idle; wait stable will frequently time out — proceed to read network anyway.

Execution Efficiency

  • Batch orchestration: write a bash script that iterates conversation ids serially in one session. For parallelism, fan out across multiple stealth browsers each with its own login.
  • Test before batch execution: run one conversation end-to-end (page 1 + at least one paginated page) first.
  • Reduce redundant pre-operations: keep the same session for many sequential conversations.
  • Error resumption: persist cursor_bottom and accumulated reply IDs per conversation after every page.
  • De-duplicate by id: the focal tweet may appear in multiple paginated responses (some endpoints repeat the focal tweet at the top of every page); merge by id.

Experience Notes

Path: {working-directory}/browser-act-skill-forge-memories/x-tweet-scraper-x-tweet-by-conversation.memory.md (working directory is determined by the Agent running the Skill, typically the project root or current working directory)

Before execution: If the file exists, read it first — it records unexpected situations encountered during past executions (e.g., a strategy has become ineffective); adjust strategy order accordingly.

After execution: If an unexpected situation is encountered (strategy became ineffective, page redesigned, anti-scraping upgraded, better path discovered), append a line: {YYYY-MM-DD}: {what happened} → {conclusion}

Normal execution does not write to the file. Do not record which conversation was scraped or how many replies were returned — those are task outputs, not experience.

More skills from browser-act/skills

  • A1688-product-detailExtracts comprehensive wholesale product data from 1688.com product detail pages: title, tiered pricing, SKU variants with dimensions/weight, product images, seller info, shop scores, buyer protection, cross-border flags, product attributes, coupon/promotion data, and review stats. Use when user mentions 1688, 1688.com, wholesale China, alibaba wholesale, B2B China sourcing, Chinese wholesale scraper, 1688 product scrape, 1688 offer, 1688 detail, extract 1688 data, pull 1688 listings, get wholesale price, 1688 supplier info, factory stats 1688, 1688 SKU variants, 1688 product attributes, 1688 shop score, DSR score 1688, 1688 buyer protection, 1688 cross-border, 1688 dropship. Also applies to: scraping bulk product data from 1688 by offer ID list, monitoring 1688 supplier metrics, extracting 1688 pricing tiers for resale analysis.
  • Aairbnb-listing-detailFetches complete Airbnb listing details for a given numeric listing ID via the internal GraphQL API, returning title, room type, description, amenities, photos, coordinates, city, house rules, highlights, ratings, review count, bedroom configuration, and property overview. Use when user mentions Airbnb listing details, Airbnb property info, Airbnb room details, get Airbnb listing data, Airbnb amenities list, Airbnb house rules, Airbnb property description, Airbnb detail page scraper, Airbnb rooms detail, Airbnb property page data, Airbnb listing info, fetch Airbnb room details, pull Airbnb listing.
  • Aairbnb-search-listingExtracts Airbnb accommodation search results from a destination query via SSR-embedded data, returning listing ID, URL, name, coordinates, rating, price, photos, and badge info for each result, plus pagination cursors for multi-page retrieval. Use when user mentions Airbnb search results, Airbnb listings, vacation rental search, short-term rental listings, scrape Airbnb, get Airbnb data, find rentals on Airbnb, Airbnb destination search, Airbnb property list, Airbnb stays search, Airbnb accommodation results, pull Airbnb listings, collect Airbnb search data, Airbnb scraper, Airbnb search page extraction, Airbnb search by destination.
  • Aamazon-alexa-qaAmazon Alexa for Shopping Q&A automation: submits questions to Amazon's Alexa/Rufus AI shopping assistant and collects response text; supports optional keyword search context (navigate to search results page before asking for category-specific answers). Use when user mentions Amazon Alexa, Rufus, Amazon shopping assistant, Amazon AI chat, ask Amazon, Amazon Q&A, automate Alexa questions, Rufus chatbot, Amazon assistant automation, collect Alexa responses, bulk question submission to Amazon, keyword search context, category research. Also applies to extracting Amazon product recommendations from conversational AI, automating repeated queries to Amazon's AI shopping feature, collecting Alexa shopping responses at scale, or market research within a specific product category.
  • Aamazon-asin-lookup-api-skillThis skill helps users extract structured product details from Amazon using a specific ASIN (Amazon Standard Identification Number). Use this skill when the user asks to get Amazon product details by ASIN, lookup Amazon product title and price using ASIN, extract Amazon product ratings and reviews count for a specific ASIN, check Amazon product availability and current price, get Amazon product description and features via ASIN, enrich product catalog with Amazon data using ASIN, monitor Amazon product price changes for specific ASINs, retrieve Amazon product brand and material information, fetch Amazon product images and specifications by ASIN, validate Amazon ASIN and get product metadata.
  • Aamazon-best-selling-products-finder-api-skillThis skill helps users extract structured best-selling product data from Amazon via the BrowserAct API. Agent should proactively apply this skill when users express needs like search for best selling products on Amazon, extract Amazon product data based on keywords, find top rated Amazon products, monitor Amazon competitor prices and sales, discover trending products on Amazon marketplace, extract Amazon product titles prices and ratings, gather Amazon product sales volume for market research, search Amazon best sellers in specific region, collect Amazon product reviews and promotion details, analyze Amazon product availability and badges, get Amazon product data for market analysis.
  • Aamazon-bestseller-listingAmazon Best Sellers listing scraper: extract product cards from any Amazon Best Sellers (zgbs) or /gp/bestsellers/ category page — returns rank (position on chart), asin, title, url, image, imageAlt, price, stars, reviewCount, ratingRaw per item, plus category metadata (categoryName, categoryFullName, categoryUrl) and pagination state (currentPage, hasNextPage, nextPageUrl). Works across all Amazon regional TLDs (amazon.com, amazon.co.uk, amazon.de, amazon.co.jp, amazon.fr, amazon.it, amazon.es, amazon.ca, amazon.com.au, amazon.in, etc.). Use when user mentions Amazon Best Sellers, Amazon bestsellers, Amazon top 100, Amazon zgbs, Amazon /zgbs/, Amazon /gp/bestsellers/, Amazon Best Sellers Rank, Amazon BSR, Amazon top ranked products, Amazon top-selling products, Amazon chart, Amazon category ranking, Amazon best sellers by category, Amazon best sellers electronics, Amazon best sellers kitchen, Amazon best sellers toys, scrape Amazon bestsellers, extract Amazon top 100, Amazon rank scraper, Amazon best seller list, Amazon leaderboard, Amazon trending products, discover trending Amazon products, Amazon niche discovery, Amazon top ranked ASINs. Also applies to competitive intelligence via ranking snapshots, spotting up-and-coming products, sourcing bestseller ASINs for further enrichment, tracking rank changes over time, and building bestseller-per-category datasets.
  • Aamazon-buy-box-monitor-api-skillThis skill helps users extract basic product details other sellers prices and seller ratings from Amazon via ASIN automatically using the BrowserAct API. Agent should proactively apply this skill when users express needs like query Amazon buy box information, monitor Amazon product prices, extract Amazon product details by ASIN, check other sellers prices on Amazon, get Amazon seller ratings and feedback count, monitor buy box ownership for a specific ASIN, track Amazon fulfillment methods for competitors, compare Amazon product prices across different sellers, retrieve Amazon buy box availability status, analyze Amazon seller profile details.
  • Aamazon-competitor-analyzerScrapes Amazon product data from ASINs using browseract.com automation API and performs surgical competitive analysis. Compares specifications, pricing, review quality, and visual strategies to identify competitor moats and vulnerabilities.
  • Aamazon-listing-competitor-analysis-skillThis skill helps users analyze Amazon competitor listings by ASIN and produce structured competitive intelligence plus strategic opportunity points for their own go-to-market. The Agent should proactively apply this skill when users want to analyze a competitor Amazon listing by ASIN, understand what a top-ranked product does right in content keywords or visuals, find market gaps and unmet buyer needs, turn competitor research into opportunity maps for their brand, identify keyword placement patterns on rival listings, extract SEO insights from Amazon product pages, reverse-engineer competitor bullet and title strategies, mine competitor reviews for buyer psychology, compare seller and A plus content patterns, run gap analysis before launching a new SKU, research why a listing wins conversion signals, synthesize whitespace you can own versus the diagnosed listing, or say just look at this ASIN with a competitive or optimization angle.
  • Aamazon-product-api-skillThis skill helps users extract structured product listings from Amazon, including titles, ASINs, prices, ratings, and specifications. Use this skill when users want to search for products on Amazon, find the best selling brand products, track price changes for items, get a list of categories with high ratings, compare different brand products on Amazon, extract Amazon product data for market research, look for products in a specific language or marketplace, analyze competitor pricing for keywords, find featured products for search terms, get technical specifications like material or color for product lists.
  • Aamazon-product-detailAmazon product detail page scraper: extract full product data from any open Amazon product detail URL (any /dp/{asin} or /gp/product/{asin} page across all Amazon regional TLDs) — returns asin, url, title, brand, price, listPrice, stars, reviewsCount, starsBreakdown (5/4/3/2/1 star percentages), answeredQuestions, inStock, inStockText, delivery, fastestDelivery, returnPolicy, breadCrumbs, features (bullet points), description, bookDescription, thumbnailImage, highResolutionImages, galleryThumbnails, productOverview (Brand/Model/etc.), attributes (tech spec table), attributesMapped (flat key-value), bestsellerRanks (rank + category + url), variantAttributes (currently selected color/size/style), variantAsins, seller (name + id + url), isAmazonChoice, amazonChoiceText, monthlyPurchaseVolume, hasAPlusContent, hasBrandStory, aiReviewsSummary, reviewsLink, productPageReviews (sample), videosCount, locationText, loadedCountryCode. Works on amazon.com, amazon.co.uk, amazon.de, amazon.co.jp, amazon.fr, amazon.it, amazon.es, amazon.ca, amazon.com.au, amazon.in, amazon.com.mx, amazon.com.br, amazon.nl, amazon.se, amazon.sg, amazon.ae, amazon.sa, amazon.pl, amazon.tr, amazon.eg. Use when user mentions Amazon product page, Amazon /dp/, Amazon dp URL, Amazon ASIN scraper, Amazon product detail, Amazon PDP, Amazon product data, Amazon product info, Amazon product fields, Amazon product attributes, Amazon full field extraction, Amazon per-ASIN enrichment, Amazon rating breakdown, Amazon stars breakdown, Amazon bestseller rank, Amazon BSR, Amazon variants, Amazon variant ASINs, Amazon color size options, Amazon feature bullets, Amazon A+ content, Amazon brand story, Amazon AI review summary, Amazon bought in past month, Amazon monthly sales volume, Amazon Amazon's Choice badge, Amazon seller info, scrape Amazon product, enrich Amazon ASIN, Amazon ASIN details, Amazon product review data. Also applies to bulk ASIN enrichment from a list of URLs, competitive product research, brand catalog audits, price and stock monitoring per ASIN, and building a normalized product dataset from a list of Amazon URLs.

All agent skills → · MCP servers