One idea shouldn't require six copy-and-paste jobs
Most social media automation starts at the wrong end of the problem.
It asks:
“How can AI generate more posts?”
We're approaching it differently at Solarc Labs.
The question we care about is:
How can one useful piece of knowledge become genuinely appropriate content for several different platforms without manually rewriting it every time?
Our current workflow looks like this:
Knowledge → AI agents → platform adaptation → Postiz → publishing
The distinction matters.
A LinkedIn post should not simply be an Instagram caption copied into another box.
A DEV.to article should not be a longer Threads post.
Different platforms have different audiences, expectations and content structures.
So the source idea stays consistent while the presentation changes.
The six-channel workflow
The first distribution layer focuses on:
- Threads
- Bluesky
- DEV.to
Instead of treating those as six independent content jobs, we treat them as six outputs from one source of truth.
The system can then adapt things such as:
- Hook
- Length
- Structure
- Technical depth
- Call to action
- Keywords
- Hashtags
- Formatting
That gives us a much more interesting automation problem than simply asking an LLM to “write a social media post.”
Why reusable knowledge matters
Content automation becomes much more useful when the system has something reliable to work from.
Rather than generating ideas from nothing every day, the content engine can work from accumulated product knowledge, research, decisions and project progress.
AI agents can then transform that context into the right output for the right channel.
The result we want is not infinite content.
It is less repetitive work and more useful reuse of knowledge.
What we're testing next
We're building the workflow in public and testing how much of the process can be automated while still keeping the output useful, specific and recognisably connected to the original idea.
That is the interesting part for us.
Not another AI caption generator.
A reusable content infrastructure layer.
One source of knowledge.
Multiple platform-native outputs.
One publishing workflow.
Top comments (0)