Merge & Tell
Turn merged GitHub PRs into social posts.
Developer Tools·Productivity·Marketing
About
Merge & Tell reads your merged GitHub pull requests and automatically writes marketing copy for social channels and changelogs. When you merge code, it uses Claude to read the diff and draft the announcement tuned to the length and voice of each connected network, helping developers share their shipped work without the awkwardness of manual writing.
The platform is built for developers and founders who want to maintain a consistent public presence without spending extra time on marketing. It filters out noise, handles small PR roundups, and lets you configure rules per repository so that routine maintenance or lockfile changes never turn into public posts.
Unlike traditional marketing tools that require manual content creation, Merge & Tell integrates directly into your existing development workflow. You simply merge your pull requests as usual, review the generated drafts, and approve them to publish across X, Bluesky, Mastodon, LinkedIn, and Discord.
Ready to launch?
Submit your product and get discovered by builders and creators worldwide.
Launch Now
FounderPlaybooks.
What other founders did to grow.
2722 dispatches from hundreds of founders, pulled from the week's best podcasts.
Apple can handle it — it is a matter of general states. If you evaluate every scenario by that rule of not punching down, you can sometimes realize hey it's okay to get a little feisty with our marketing when it comes to competing with a company like Apple.
It's OK to be feisty with giants — punching up is different from punching down
When a small app calls out Apple Weather for going down, that's punching up — the giant has infinite resources to fix it and can absorb criticism. When an app gloats over a small competitor's shutdown or a category of people losing jobs, that's punching down — and writers, audiences, and even potential customers will notice. The rule: feisty is fine at scale, mean-spirited reads badly at any scale.
We went on Craigslist and started saying 'hey, we're looking for people to weigh their plants every day twice per day for a couple months.' We had hundreds of responses. People thought it was cool to be doing citizen science — we ended up with people in Berlin and Sydney.
Crowdsource training data on Craigslist — citizen-science weight-tracking
Greg needed global plant water-loss data to train its ML model. Rather than build a warehouse, Alex hit Craigslist and got hundreds of volunteers — including across hemispheres (which mattered for solar-radiation modeling). The product's ML core was trained on data they didn't pay for, contributed by people who liked the mission.
There's a play for whatever you're stuck on.
Read all 2722 playbooks