Claude Social Media Prompts for 2026 (Copy-Paste Library)
Claude Social Media Prompts for 2026 (Copy-Paste Library)
When the brief is “sound exactly like us, don’t invent claims, and make the LinkedIn post actually teach something,” Claude is usually the better drafting partner than a speed-first chat. Long context lets you load voice guides, past winners, and compliance notes once — then ask for careful rewrites instead of roulette drafts.
Full index: AI social media prompts library. Product path: Agent Mode → Calendar Mode.
Key Takeaways
- Claude excels at long-context brand voice and fidelity across a campaign
- Use it for careful rewrites, LinkedIn depth, and agency compliance reviews
- Projects hold guidelines; Artifacts are ideal for calendars and multi-post packs
- Draft in Claude, approve claims, then publish from Postevery’s calendar
Why Claude for social
Claude’s edge is restraint. It follows “do not claim X” instructions more reliably than models optimized for chatty volume. That matters for regulated clients, B2B thought leadership, and any brand that already has a voice document people ignore when they are rushing.
Use Claude when:
- You are rewriting a human draft to tighten voice, not inventing from a blank page.
- LinkedIn needs 180–250 words with a real teaching structure.
- An agency needs a compliance pass before the client sees the calendar.
- You can load several sample posts + guidelines into a Project and reuse them all month.
Use something else when you need 20 hook variants in 30 seconds (ChatGPT) or live-cited trend research (Perplexity / Gemini).
How to brief Claude well
Claude prefers complete briefs over rapid ping-pong. Front-load: audience, voice samples, forbidden claims, platform, length, and what “done” looks like. Ask it to flag uncertainty instead of guessing.
- Create or open a Project for the brand.
- Upload voice samples and a short claims policy.
- Paste the context block, then one task with an explicit structure.
- For multi-asset work, ask for an Artifact (calendar table, carousel outline, reply kit) so you can edit in place.
Context block (paste first)
Slightly longer than a ChatGPT block on purpose — Claude uses the extra voice and compliance detail:
You are the senior social editor for [brand]. Prefer careful, on-brand copy over cleverness.
Brand: [name] — [one-liner] Audience: [who, role, pain] Voice: [3 adjectives]. Match these samples closely:
- [sample post 1]
- [sample post 2] Offer / CTA rules: [what we can and cannot promote] Forbidden claims: [list]. If a draft would require a claim you cannot verify, flag it instead of inventing. Platforms this week: [list] Goal: [metric] Default length: LinkedIn 180–220 words; Instagram 80–120; X under 240 characters unless I say otherwise.
Projects and Artifacts tips
Projects: One Project per client or brand. Keep the context block, compliance notes, and 5–10 approved posts as Project knowledge. Start weekly batches there so Claude does not drift toward generic LinkedIn-speak.
Artifacts: Ask Claude to put calendars, multi-platform adaptations, and carousel outlines into an Artifact. Edit the table in place, then export into Postevery. For agency reviews, an Artifact with a “claim risk” column saves a round of Slack.
Extended thinking (when available): Use it for compliance rewrites and “make this sound like our CEO without inventing anecdotes” — not for simple hook lists.
Do
- Load 2 voice samples and a forbidden-claims list before any campaign draft
- Ask Claude to quote which guideline it followed when rewriting
- Use Artifacts for calendars and multi-post packs you will edit
Don't
- Ask for 20 wild hook variants when you need one careful LinkedIn post
- Let Claude invent customer stories — provide the anecdote or forbid fiction
- Skip a human compliance skim on regulated industries
Prompt library (12+)
Rewrite the following to match our voice samples in the Project. Keep every factual claim. Do not add new metrics or testimonials. Goals: shorter sentences, one concrete example, remove buzzwords. Flag any line you could not verify. Draft: [paste]
Write a LinkedIn post (180–220 words) teaching [insight] for [audience]. Structure: hook (no question cliché), short lived experience or observed pattern, 3 actionable bullets, CTA inviting a comment with a specific prompt. No external link in body. Do not invent a personal story — use only: [facts / anecdote].
Produce 4 captions for [platform] about [topic] for client [name]. Compliance notes: [rules]. For each caption add: claim risk (low/med/high) + one safer alternative line if risk is med/high. Tone: [voice]. Soft CTA only.
For client [name] ([industry]), create 3 educational posts and 1 proof post across [platforms]. Return a review table: platform | hook | caption | asset note | claim risk | editor note for human reviewer. Use only approved proof points: [list].
Write 8 hooks for [platform] on [topic]. Max 12 words. Each must include a concrete noun or number. After the list, recommend the top 3 for [audience] with a one-sentence rationale each. No "Are you tired of" patterns.
Adapt this core post for LinkedIn, Instagram, X, and TikTok caption. Preserve meaning; change hook, length, and CTA. Call out anything that would be off-brand if copied verbatim. Core: [paste] Voice rules: [summary]
Build a 7-day content calendar for [brand] on [platforms] as a table Artifact. Columns: day, platform, format, hook, caption outline, asset, CTA, claim risk. Mix: 40% educational, 30% proof, 20% product, 10% personality. No invented case studies.
Ghostwrite a LinkedIn post in the voice of [name/role] about [topic]. Voice notes: [cadence, phrases they use, phrases they never use]. Do not invent biography. Use only these talking points: [list]. Length: 150–200 words. End with a question that invites peer debate, not a product pitch.
Write a 6-post LinkedIn or X thread on [topic]. Each post must stand alone if skimmed; together they teach one framework. No cliffhanger spam. Soft CTA only on the final post. Voice: [voice].
This draft sounds either AI-generic or overly corporate. Rewrite for human specificity while staying professional. Keep all facts. Replace abstractions with verbs and objects. Max one metaphor. Text: [paste]
Draft 10 reply templates for comments on [brand]'s posts about [sensitive topic]. Categories: supportive, skeptical, off-topic, and potential compliance risk. Stay calm, on-brand, and never invent policy. Escalate rule: [when to hand to human].
Outline an 8-slide LinkedIn or Instagram carousel teaching [framework]. Per slide: headline, body (≤30 words), visual direction. Slide 1 = tension. Final = soft CTA. Ensure claims match: [approved points]. Flag any slide that needs a source.
Compare Draft A and Draft B. Produce an edit memo: what to keep from each, a merged final post, and 3 residual risks for legal/brand review. A: [paste] B: [paste]
Ship with Postevery
Brief inside a Project
Review claims
Export the pack
Schedule in Postevery
Also see: ChatGPT prompts, Best AI social media schedulers 2026.
Frequently Asked Questions
Are these Claude prompts free to use?
Yes. Copy and use them in Claude chats, Projects, or Artifacts workflows.
When should I choose Claude over ChatGPT for social?
Choose Claude when brand voice, LinkedIn depth, careful rewrites, or compliance matter more than raw iteration speed.
How do Projects help with brand voice?
A Project keeps guidelines, samples, and forbidden claims loaded so you are not re-pasting a style guide every Monday.
How do I use Claude with Postevery Agent Mode?
Generate and review in Claude, then connect your agent in Postevery Agent Mode and schedule in Calendar Mode. Details are in the Agent Mode guide.
Where is the master library?
https://postevery.com/blog/ai-social-media-prompts-library — with sibling packs for ChatGPT, Gemini, Copilot, and Perplexity.
One calendar. 27+ platforms. AI built in.
Schedule with Calendar Mode or automate with Agent Mode. Start a 7-day free trial on any monthly plan.
Start free trial