agentleFS
moses607moses607/socialforge/skills/engagement-booster/SKILL.md

engagement-booster

Skill3 starsChanged 4 months agoMITView on GitHub

Engineers the engagement signal algorithms reward — comments, replies, saves, shares, and DMs — instead of chasing raw reach. Produces a per-post engagement playbook plus reusable reply and DM templates covering first-hour tactics, reply strategies that grow reach, low-friction CTAs, and community rituals. Use when the user wants more comments, a reply strategy, a DM flow, or community-building. Trigger on engagement, comments, replies, DMs, community, "boost engagement", "get more comments", "reply strategy". Works with any capable model.

The file

Files
SKILL.md

Engagement Booster

Algorithms rank content by the engagement it provokes in the first minutes, and every reply, save, and DM is a vote that widens distribution — so treat each post as the opening move of a conversation, not a broadcast. The highest-leverage work happens AFTER you hit publish: the first hour of replies, the discussion you deliberately seed, and the follow-up in DMs. Optimize for comments and saves (the strongest signals) over likes, engineer low-friction ways for people to participate, and turn one-time engagers into a community through repeatable rituals.

1. First-hour tactics (the golden window)

  1. Be present at publish. Post when your audience is active, then stay for 60 minutes. Early velocity of comments is what triggers the algorithm to expand reach.
  2. Reply to every comment within minutes, and make replies substantive — a question back, not just "thanks." Each reply is a fresh comment that boosts the count and pulls the commenter back for a second interaction.
  3. Seed a discussion-starter comment yourself in the first minute: pin a spicy question, a "hot take: ___, agree?", or the #1 objection people will have. This gives lurkers a low-stakes entry point.
  4. Bake comment-bait into the post. End with a specific, easy question ("A or B?", "what's your #1 ___?") or a fill-in-the-blank. Vague "thoughts?" underperforms concrete binary/list prompts.

2. Reply strategies that grow reach

  1. Add value, don't just acknowledge. Answer with a bonus tip or resource so the reply is worth reading — screenshotting-worthy replies get shared.
  2. Always end with a follow-up question to keep the thread alive; longer threads = more signal.
  3. Pin your best comment (a great question or your seeded starter) to steer the conversation and lift its visibility.
  4. Reward good comments publicly — heart them, reply with genuine praise, or feature them. This trains your audience that commenting gets noticed, so they do it more.
  5. Turn objections into content. A recurring question in replies is your next post; tell them so ("great Q — full post coming").

3. DM flows & low-friction CTAs

  1. Welcome DM: thank new followers, give one piece of value, ask one light question. NO pitch. Goal is a reply, which opens the messaging relationship.
  2. Keyword-to-DM: "Comment WORD and I'll send you the [guide]." This spikes comments AND opens DMs at scale. Deliver value first; soft offer only after 2–3 friendly exchanges.
  3. Participation mechanics: polls, "this or that", quizzes, and "rate 1–10" ask for a tap, not an essay — the lowest-friction engagement. Rotate them in weekly.
  4. CTA ladder by effort: like → save → poll tap → comment a word → thoughtful comment → share → DM. Ask for one rung, not the whole ladder.

Output template

POST: <topic/link>
IN-POST COMMENT-BAIT: "<specific question or fill-in-the-blank>"
SEEDED FIRST COMMENT (post yourself, minute 1): "<hot take / question>"
FIRST-HOUR CHECKLIST:
  - [ ] Reply to every comment <5 min, each ends with a question
  - [ ] Pin best comment: "<which>"
  - [ ] Heart + reply to top 3 commenters
CTA THIS POST: "<one ask — e.g. Comment 'GROW' for the checklist>"
REPLY TEMPLATES:
  - Value add: "Great point — plus [bonus tip]. Have you tried [X]?"
  - Objection: "Fair! The reason is [why]. What's stopped you so far?"
  - Praise: "This comment is gold 🙌 [restate their insight]."
DM FLOW:
  - Welcome: "Thanks for following! [one tip]. What are you working on right now?"
  - Keyword: "Sent! While you're here — what's your biggest struggle with [topic]?"
COMMUNITY RITUAL: <weekly series / recurring format / shout-out day>

Platform variants

  • Instagram/Threads: keyword-to-DM automation, polls/questions in Stories, reply to comments with a Reel.
  • LinkedIn: thoughtful questions + reply to every comment in the first hour; dwell time and comments drive reach. Avoid outbound links in-post.
  • X/Twitter: reply speed and quote-replies; ask binary questions; the reply guy strategy under bigger accounts.
  • YouTube: pin a question comment, heart early comments, reply to build a thread under the video.
  • TikTok: reply to comments WITH a video (huge reach unlock); ask a question in the caption.

Rules

  • Engineer for comments, saves, and shares — not likes. They are the signals that move distribution.
  • Show up in the first hour or don't post; the golden window is non-negotiable.
  • Reply to every early comment, and every reply ends with a question. No dead-end "thanks."
  • Give value in DMs before any ask — never pitch on the first message.
  • One clear CTA per post, matched to the lowest-friction rung the audience will actually do.
  • Build rituals (series, recurring formats, shout-outs) so engagement compounds instead of resetting each post.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.