Threads User Posts Skill for Cursor

Threads — User Posts

> username → list of public posts with engagement metrics (SSR + scroll pagination)

Language

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

Objective

Extract public posts from a given Threads user's profile page using SSR-embedded JSON data, with optional scroll-triggered pagination to load more posts.

Prerequisites

• A browser is open and connected via browser-act
• Login recommended: Threads enforces a login wall on profile pages in most regions as of mid-2026. A logged-in browser session is required for reliable access. Unauthenticated access via US-based proxies may work in some configurations but is not guaranteed.

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 Threads has been confirmed in the current session → skip this step.

Otherwise: navigate https://www.threads.com and check page state:

• User avatar or username visible at top → logged in, continue
• "Log in" / "Sign up" buttons visible → not logged in, inform the user that login is needed, assist completing the login flow via the Threads login page

If not logged in and user cannot log in → the capability may still work on some proxy/IP configurations; attempt navigate https://www.threads.com/@{username} and check if profile content loads. If the page redirects to /login/, terminate and inform the user that a logged-in session is required.

Capability Components

> This Skill's operational boundary = what the user can manually do in their browser. It only reads data already displayed to the user on the page. JS code is encapsulated in Python files under the scripts/ directory, invoked via eval "$(python scripts/xxx.py {params})". $(...) is bash syntax; it is recommended to use the bash tool for execution.

SSR: Extract posts from profile page (initial load)

Navigate to the user's profile page first, then extract all posts embedded in the initial page HTML:

• navigate https://www.threads.com/@{username}
• wait stable
• eval "$(python scripts/extract-posts.py)"

Output example:

{
  "posts": [
    {
      "id": "3936653356768062022",          // post internal ID (pk)
      "code": "DKmzN8wJzFG",               // post short code for URL
      "url": "https://www.threads.com/@zuck/post/DKmzN8wJzFG",  // direct post link
      "text": "Excited to share...",        // post text content, null if no caption
      "taken_at": 1780953646,              // Unix timestamp of post creation
      "like_count": 3583,                  // number of likes
      "reply_count": 3039,                 // number of direct replies
      "repost_count": 232,                 // number of reposts
      "quote_count": 114,                  // number of quote posts
      "reshare_count": 460,                // number of reshares
      "is_reply": false,                   // true if this post is a reply to another
      "media_type": 19,                    // 1=photo, 2=video, 8=carousel, 19=text-only
      "has_media": false,                  // true if post contains image/video/carousel
      "user": {
        "pk": "63055343223",              // Threads-specific user ID
        "username": "zuck",
        "full_name": "Mark Zuckerberg",
        "is_verified": true
      }
    }
  ],
  "count": 4,                             // number of posts in this batch
  "page_info": {
    "end_cursor": "QVFES...",             // cursor for next page (null when no more)
    "has_next_page": true,                // whether more posts are available
    "has_previous_page": false,
    "start_cursor": null
  }
}

Error handling: If error: true is returned, check that the profile page was fully loaded (navigate and wait stable completed without timeout). If mediaData not found, the page may have redirected to a login wall or the username is invalid.

DOM: Scroll to trigger more posts

When page_info.has_next_page is true, scroll the page to trigger GraphQL auto-load:

eval "$(python scripts/scroll-load-more.py)"

Then wait stable and read the new batch from network traffic:

network requests --type xhr,fetch --filter threads.com

Find the POST https://www.threads.com/graphql/query request(s) that appeared after the scroll. Read the response:

network request <id>

The response body contains data.mediaData.edges[] with the same post structure as the SSR extraction output. Each edge has node.thread_items[0].post with the same field layout (id/pk, code, caption.text, taken_at, like_count, text_post_app_info.direct_reply_count, etc.).

Error handling: If no new graphql/query requests appear after scrolling, the user may have reached the login wall (~15 posts without authentication) or all posts have loaded. Check for a "Log in to see more" message on screen via screenshot.

Composite: Full profile post collection (SSR + scroll pagination)

To collect all available posts for a user:

• navigate https://www.threads.com/@{username} → wait stable
• eval "$(python scripts/extract-posts.py)" → collect initial posts, note page_info
• While page_info.has_next_page == true:

a. eval "$(python scripts/scroll-load-more.py)"

b. wait stable

c. network requests --type xhr,fetch --filter threads.com → locate new graphql/query POST

d. network request <id> → extract data.mediaData.edges[] posts

e. Check data.mediaData.page_info.has_next_page in response to decide whether to continue

• Stop when has_next_page == false or login wall encountered (~15 posts without login)

Pagination

DOM Pagination: Scroll #scrollview via eval "$(python scripts/scroll-load-more.py)", then wait stable and read new POST /graphql/query response from network traffic. Termination: page_info.has_next_page == false in the GraphQL response, or login wall visible on screen (approximately 15 posts without authentication).

Success Criteria

result.count >= 1 and result.posts[0].id != null and result.posts[0].user.username != null

Known Limitations

• Login wall: As of mid-2026, Threads redirects unauthenticated profile page requests to the login page in most regions. A logged-in browser session is required for reliable use; unauthenticated access works only on specific proxy/IP configurations.
• Authenticated access: up to approximately 15 posts per profile before hitting a soft pagination limit; further posts require additional scroll cycles.
• Private accounts: profile page shows no posts even for logged-in users who don't follow the account.
• Deleted or suspended accounts return mediaData not found error.
• taken_at is a Unix timestamp; convert to readable date with new Date(taken_at * 1000).toISOString().

Execution Efficiency

• Batch orchestration: Write a bash script to loop through the command templates serially within a single session; do not parallelize within one browser. Add 2-3 second intervals between profile navigations to avoid rate limiting. For higher throughput, distribute usernames across multiple parallel browser sessions.
• Test before batch execution: After writing a batch script, test with 1-2 usernames first to verify the script runs correctly; only then run the full batch.
• Reduce redundant pre-operations: When collecting multiple profiles in the same session, the session state is preserved across navigations — no need to reinitialize.
• Error resumption: Save results per username during batch processing; on failure, resume from the breakpoint rather than starting over.

Experience Notes

Path: {working-directory}/browser-act-skill-forge-memories/threads-scraper-threads-user-posts.memory.md

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 what keywords were used or how many results were returned — those are task outputs, not experience.