The problem
Staying consistent across channels is a full-time job
Every platform wants a different format, length, and tone. Reformatting one idea for LinkedIn, X, Instagram, and a newsletter — then scheduling and monitoring replies — eats hours a day. Skip it and the account goes quiet; rush it and the brand voice drifts.
Autopilot repurposes once, adapts per platform, and keeps the calendar full without going off-brand.
Primary objective
One source asset → a full, on-brand, multi-platform week — scheduled and monitored.
How it works
Ingest → repurpose → schedule → engage
Ingest
Blog, video, PDF or a raw idea
→
Repurpose
Platform-native variants, on voice
→
Schedule
Best-time queue, human-approved
→
Engage
Reply drafts for comments & DMs
What it does
- ✓ One-to-many repurposing
- ✓ Per-platform formatting & length
- ✓ Hashtag & hook optimisation
- ✓ Best-time scheduling
- ✓ Comment / DM reply drafts
- ✓ Performance recap & reuse of winners
Built with
- ✓ Claude for adaptation & replies
- ✓ Brand-voice profile per client
- ✓ LinkedIn, X, Instagram, Facebook APIs
- ✓ Scheduler with approval queue
- ✓ Analytics ingestion for feedback loop
Guardrails
- ✓ Nothing posts without approval
- ✓ Brand-safety & tone filters
- ✓ Claim & fact checks on posts
- ✓ Escalation for sensitive replies
- ✓ Per-platform policy compliance
Under the hood
One asset, many platform-native variants
Each platform gets its own constraints and voice — same idea, native execution — and every post lands in the approval queue first.
social/repurpose.py
PYTHON
from anthropic import Anthropic client = Anthropic() PLATFORMS = { "linkedin": {"limit": 3000, "tone": "professional, story-led"}, "x": {"limit": 280, "tone": "punchy, one hook"}, "instagram": {"limit": 2200, "tone": "warm, visual, emoji-ok"}, } def repurpose(asset: str, brand_voice: str) -> dict: posts = {} for name, cfg in PLATFORMS.items(): prompt = f"""Rewrite this asset as a native {name} post. Max {cfg['limit']} chars. Tone: {cfg['tone']}. Brand voice: {brand_voice}. Keep the core idea; adapt the format. Source: {asset}""" msg = client.messages.create( model="claude-sonnet-4-20250514", max_tokens=700, temperature=0.6, messages=[{"role": "user", "content": prompt}], ) posts[name] = msg.content[0].text return posts # -> approval queue, then scheduler
Outcomes
1→20+
Posts per source asset
Hours→min
To fill the calendar
On-brand
Voice held across channels
Always-on
Consistent presence
Keep every channel alive on autopilot
Great with the Content Engine — every article becomes a week of social.
More SEO & Marketing solutions