Prompt Generator
ClaudeTemplate, 7 blanks to fill

You are a social media writer planning a week of posts for a product launch. <context> Product: [PRODUCT] What it does in one sentence: [ONE LINER] Who it is for: [AUDIENCE] Launch date: [LAUNCH DATE] Platform: [PLATFORM] Proof we can show (screenshots, a demo clip, quotes we have permission to use): [PROOF] Link: [LINK] </context> <instructions> 1. Plan seven posts, one per day, starting two days before the launch date. 2. Give each day one job, in this order: name the problem, show how it was made, announce, demo, answer an objection, share the proof, last call. 3. Write each post for the platform: follow its usual length, line breaks and hashtag habits. 4. Open every post with a first line that works on its own in the feed. 5. Put the link only in posts where a click makes sense, and say where it goes: in the post, the first comment or the bio. </instructions> <constraints> - Use only the proof listed in the context. Never invent users, numbers, reviews or press mentions. - No two posts may open the same way. - Use emoji only if people on this platform expect them, and at most one per post. - Each post asks for one action at most. </constraints> <output_format> For each day: the day and date, the job of the post, the post text, and one line on the visual to attach. After the seven posts: two spare opening lines to swap in if a post gets little reach. </output_format>

The words in brackets are the parts you replace with your own details.

What each part of the prompt does

One job per day
Seven posts that all say "we launched" wear out an audience by day three. Giving each day a fixed job (problem, making-of, announce, demo, objection, proof, last call) makes every post worth reading on its own.
Facts inside the context tag
Claude keeps closely to what sits inside <context>. Putting the one-liner, proof and link there means every post repeats the same accurate facts instead of paraphrasing them into new claims.
Proof you actually have
The constraint against invented users, numbers or press is the most important line. Launch posts are where made-up social proof slips in, and it is also where people check.
Spare openings
Two extra first lines at the end give you something to swap in when a post gets little reach, without rewriting the whole post under time pressure on launch day.

Variations to try

Same idea, one thing changed.

  • Three-post short series

    You are a social media writer planning a week of posts for a product launch. <context> Product: [PRODUCT] What it does in one sentence: [ONE LINER] Who it is for: [AUDIENCE] Launch date: [LAUNCH DATE] Platform: [PLATFORM] Proof we can show (screenshots, a demo clip, quotes we have permission to use): [PROOF] Link: [LINK] </context> <instructions> 1. Plan three posts: one the day before the launch, one on launch day and one two days after. 2. Give each post one job, in this order: name the problem, announce with a demo, share the proof. 3. Write each post for the platform: follow its usual length, line breaks and hashtag habits. 4. Open every post with a first line that works on its own in the feed. 5. Put the link only in posts where a click makes sense, and say where it goes: in the post, the first comment or the bio. </instructions> <constraints> - Use only the proof listed in the context. Never invent users, numbers, reviews or press mentions. - No two posts may open the same way. - Use emoji only if people on this platform expect them, and at most one per post. - Each post asks for one action at most. </constraints> <output_format> For each day: the day and date, the job of the post, the post text, and one line on the visual to attach. After the three posts: two spare opening lines to swap in if a post gets little reach. </output_format>

  • Cross-platform set

    You are a social media writer planning a week of posts for a product launch. <context> Product: [PRODUCT] What it does in one sentence: [ONE LINER] Who it is for: [AUDIENCE] Launch date: [LAUNCH DATE] Platforms: [PLATFORM 1] and [PLATFORM 2] Proof we can show (screenshots, a demo clip, quotes we have permission to use): [PROOF] Link: [LINK] </context> <instructions> 1. Plan seven posts, one per day, starting two days before the launch date. 2. Give each day one job, in this order: name the problem, show how it was made, announce, demo, answer an objection, share the proof, last call. 3. Write each post twice, once for each platform, following that platform's usual length, line breaks and hashtag habits. 4. Open every post with a first line that works on its own in the feed. 5. Put the link only in posts where a click makes sense, and say where it goes: in the post, the first comment or the bio. </instructions> <constraints> - Use only the proof listed in the context. Never invent users, numbers, reviews or press mentions. - No two posts may open the same way. - Use emoji only if people on this platform expect them, and at most one per post. - Each post asks for one action at most. </constraints> <output_format> For each day: the day and date, the job of the post, the post text, and one line on the visual to attach. After the seven posts: two spare opening lines to swap in if a post gets little reach. </output_format>

If the result isn't right

  • Run the prompt once per platform; a single plan that tries to fit every network ends up sounding right on none of them.
  • If you have no proof yet, write "none yet" in [PROOF] and Claude will shift day six to a behind-the-scenes post instead.
  • Ask for the visual notes as a shot list and you can capture all seven screenshots or clips in one sitting.