This week: safer post-now gates and clearer LinkedIn dry-run labels

This week I focused on two things that matter if you use git-to-market to turn commits into public posts: making "post now" behavior safer, and making LinkedIn dry-runs honest and obvious. Both changes reduce accidental publishing and make testing feel trustworthy.

What shipped

• –Post-now gates: before we would sometimes push a queued social post immediately after a commit if settings changed. Now there are clear, deliberate gates so a post marked "post now" only goes live after you confirm the new intent. That prevents unexpected posts when you tweak scheduling or account settings.
• –LinkedIn dry-run truthfulness: LinkedIn dry-runs previously looked almost identical to a live post, which led people to assume the draft was published. Dry-runs now show an explicit, unmistakable banner and a slight content watermark so you can tell at a glance it’s a preview, not live. The content generation is the same; only the presentation changed to reduce mistakes.
• –Error surfacing: we improved how failures are reported back to you. If a post fails to publish (account disconnected, API limit, content block), you now get a concise error message and a suggested next step instead of a generic failure note.

Why it matters

People push code to GitHub multiple times a day. If your marketing tool posts automatically without clear confirmations, one small tweak can create an embarrassing public post. These changes reduce that risk. You keep the convenience of automatic posting while gaining control and clarity about what actually goes live.

Clear dry-runs mean you can test tone and formatting for LinkedIn without worrying that a preview escaped. Better error messages mean less time digging through logs or guesswork when something doesn’t publish. Together these changes lower the friction for building in public consistently.

What's next

I’ll watch how these gates affect daily flows. If people find the confirmation step too interruptive for high-frequency committers, I’ll add a toggle for a trusted mode that remembers your intent for a short period. I’m also planning one follow-up: a small UI change that surfaces the last successful publish timestamp for each channel so you can see at a glance when something actually went out.

Related reading

[This week: post-now gates and clearer LinkedIn dry-run truthfulness]()

[This week: post-now gates and clearer LinkedIn dry-run truthfulness]()

[This week: post-now gates and clearer LinkedIn dry-run truthfulness]()