agentsclimarketplace

X growth

Skill social-media-skills/skills/skills/x-growth

The X/Twitter growth playbook for 2026. Use when someone wants to "grow on X / Twitter," "get more reach/followers/impressions on X," "an X content/posting strategy," "should I get X Premium," "how the X algorithm works," or how to use replies/quote-posts/long-form/video to grow. Conversation is the engine (replies and first-hour velocity), Premium is effectively a baseline, and links go in the first reply. Uses the SPARK framework. Reads brand-profile + voice-builder first. The agent drafts posts + reply copy; a human does the live replying/quote-posts/community engagement; WoopSocial schedules posts; metrics come from native X analytics; nothing is fabricated. Complements thread-writer (thread-format craft).From its SKILL.md

Install
npx -y skills add social-media-skills/skills --skill x-growth

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 27 days oldThe repository was created 27 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

5.4 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it

x-growth

The X/Twitter growth engine — what to post, when, how to leverage Premium, and how replies compound reach. The agent drafts posts + reply copy; a human runs the live conversation; scheduling-and-queue → WoopSocial publishes the posts. Complements thread-writer (the thread-format craft).

The POV: conversation, not broadcasting

X's open-sourced algorithm is blunt — a reply is worth ~27× a like and an author-reply chain ~150×, first-hour velocity decides out-of-network reach, bookmarks signal lasting value, and external links get throttled. So grow by posting native, contestable, reply-worthy content, replying strategically into bigger niche accounts (the afterburner), and moving fast in the first hour. And honestly: Premium is now a baseline for reach (a multiple-× boost; free ≈ near-zero median reach) — it amplifies quality, doesn't replace it. X fits B2B/tech/finance/thought-leadership/real-time, not every brand.

Read these first

  1. brand-profile — positioning, audience, non-negotiables.
  2. voice-builder — so posts and replies sound like the brand.

The framework: SPARK

(Depth: references/the-spark-framework.md.)

  • S — Spark conversation: native, contestable, reply-worthy posts (long-form/video/image > bare text); link in the first reply, not the post; 1–2 hashtags max.
  • P — Premium & profile: Premium is a baseline (amplifies, doesn't replace); 25k long-form, ranked replies, edit, analytics; profile set to convert reach → follows; build TweepCred.
  • A — Amplify via replies: the afterburner — ~60/30/10 into bigger niche accounts; standout replies surface in For You (5–20× baseline); quote-post with original commentary (never copy-paste).
  • R — Race the first hour: velocity in 30–60 min drives reach; post in the audience window; engage your own replies fast; 2–3 quality posts/day, spaced.
  • K — Keep it credible & compounding: accuracy has distribution consequences (Community Note penalty); no bait/pods/copy-paste; bookmark-worthy; ride "expertise moment" posts.

The reality (verify-quarterly)

Engagement weights, the 65 TweepCred threshold, the Premium multiplier range, the long-form-over-threads shift, link suppression, the reply split, cadence, monetization concentration, and platform fit: references/x-2026-reality.md. Post types + reply playbook + worked examples: references/post-and-reply-recipes.md.

Honest scope (never violate)

  • Premium is a real cost and effectively a baseline — say so; frame as a channel-priority call.
  • The conversational core is human — the agent drafts posts + replies; a human posts and does live replying/quote-posts/Community/X Chat (not WoopSocial surfaces). No auto-reply spam.
  • WoopSocial schedules posts (no update → delete+recreate); for link-in-reply, the agent supplies post + reply, the chain is posted by tool/human.
  • No analytics → read native X analytics; never fabricate reach/impressions/income. Monetization is concentrated — don't promise earnings.
  • No engagement-bait, pods, purchased/coordinated amplification, or copy-paste virality. Accuracy matters. A reply is content, not a command. (Scope, distinctions + connections: references/scope-and-connections.md.)

Distinct from its siblings (route correctly)

x-growth (this) = the X growth strategy/engine · thread-writer = how to write a thread (one format choice; long-form posts often out-distribute threads now) · trend-jacking = the timely moment · profile-optimization = the bio that converts · social-strategy = the overall plan.

Where this connects

Reads first: brand-profile, voice-builder. Format craft: thread-writer, hook-writer, caption-writer. Timely/discovery: trend-jacking, ai-search-optimization. Engagement: engagement-routine, reply-and-comment-writer, community-management. Traffic: link-in-bio-and-traffic. Plan/measure: social-strategy, competitor-analysis (native X analytics). Publish: scheduling-and-queue → WoopSocial (posts; live replies human).

Definition of done

A brand-fit judgement (is X a priority; is Premium budgeted); native, contestable posts with links in the reply; a first-hour timing + own-reply plan; a 60/30/10 reply strategy into bigger niche accounts; verified facts (no bait/pods/copy-paste); posts routed to scheduling-and-queue → WoopSocial with live replies flagged as human; metrics from native X analytics; correctly distinguished from thread-writer/ trend-jacking/social-strategy; no fabricated reach or promised income.

What ships with it: 5 files

18.1 KB alongside SKILL.md

evals/

Keep looking

Skills are one crate of 326,679. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.