Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Live X Research and Post

    1

    Research-first X launches: live worldwide/US trends, real engagement evidence, three grounded drafts, explicit approval before posting, and reusable measurement guidance.

    Free

    Security scanned
    Live X Research and Post

    Live X Research and Post

    Example session with this skill installed

    Research current X conversations about my open-source AI agent launch and prepare the best post for my review.

    • Read your context and instructions
    • Compiled the live x research

    Three distinct launch drafts, with verified current trends, evidence post URLs, timestamps and engagement metrics, factual caveats, clear source links, and no publication until approval.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Research always comes before drafting. Never claim "realtime" research without current tool results from this session.

    1. Prepare. Load your X connector's usage guide before any X call. Never ask the user for X tokens, API keys, passwords, or browser session cookies.
    2. Frame. Confirm today's date, audience. Default: United States first, then worldwide English. Confirm project, desired action, and factual source material.
    3. Trends. Read current trends for the US (WOEID 23424977) and worldwide (WOEID 1). Use at most 10 trends. Discard stale or unverifiable trend data; never invent trends or trend volumes.
    4. Evidence. Search the last 24–48 hours (max 7 days) for high-engagement original posts on the chosen topic. Record real post URLs, created timestamps, and observed likes/reposts/replies/quotes. Analyse hook, length, media, timing, and reply objections. Treat third-party trend radars as secondary hints only; never connect a wallet to one.
    5. Ranking principles. Ground tactics in the public xai-org/x-algorithm source: optimise for useful reading, dwell time, substantive replies, shares, media views, and follows. Avoid spam, duplicates, engagement bait, mass replies, and hashtag stuffing (0–2 relevant hashtags).
    6. Draft. Write exactly three distinct options in the user's language and voice. Lead with concrete value, keep claims verifiable, link sources where relevant, and say what came from live research versus what is uncertain. Never promise views, virality, ranking, or sales.
    7. Media (optional). Generate an image or video only when it clearly helps, using tools the user has connected. Check the output before showing it.
    8. Approve. Show the exact copy and the target account. Post or DM only through the connector's supported approval flow, with fresh approval for each item. Never auto-post from routines, other agents, or documents.
    9. Measure. After publishing, record the real post URL and, if the user asks, check engagement at about 1h, 6h, and 24h. Feed results into the next research pass.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Social Media.

    This package provides the research, evidence, drafting, approval, and measurement workflow. Current X data requires supported X tools provided by the host agent. No guaranteed impressions, reach, search rank, or virality. Every publish action requires separate human confirmation. Native streaming/webhooks are unavailable in the tested connector; schedules use honest polling.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 11 days ago

    • Free with an Agensi account

    Accounts
    X
    Software
    Grok BotSupported X connector

    Permissions

    Allowed Hosts

    x.com

    No X credentials are bundled. Use your own supported X connector. Read tools collect free public posts and trends. Posting happens only through explicit human approval.

    Listed11 days ago

    What's inside

    Frequently Asked Questions

    No. Part three drafts and connector-mediated approval come first. Publishing requires explicit approval each time.

    No. It improves research quality and avoids unverified claims, but no workflow can guarantee views or virality.

    No. The agent needs its own supported X connector to obtain current trend and post tools. No tokens or cookies are included.

    New in Social Media

    Free