Build / Replit content feed app
How to build a content feed app with Replit.
Use Replit to build the app and Synorb REST to supply fresh, source-grounded content. The important part is simple: browser code calls your Replit backend, and the backend calls Synorb with server-side credentials.
Architecture
Keep credentials in Replit secrets.
Frontend
Feed UI
Calls/api/feed in your Replit app
Receivesnormalized rows, pagination, usage state
Backend
Synorb REST
SecretsSYNORB_API_KEY, SYNORB_SECRET
Returnssource-linked Manifests as content feed rows
Build prompt
Paste this into Replit.
Build a content feed app powered by Synorb REST. Create a backend route /api/feed that reads SYNORB_API_KEY and SYNORB_SECRET from Replit secrets, calls Synorb REST for recent Manifests, and returns normalized rows: id, headline, summary, publishedDate, sourceUrl, sourceName, tags. Do not expose Synorb credentials to the browser. Build a frontend with topic filters, latest grid, source links, loading state, empty state, error state, and Load more using pagination.
FAQ
Builder notes.
Should Replit use MCP for production?
No. The deployed app should use REST from a backend route. Use MCP separately if an agent needs to explore Streams while you design the app.
What does the frontend receive?
A small normalized JSON shape: rows, pagination, usage, and maybe filter metadata. Credentials stay server-side.
Which Synorb page should I read next?
Read /structured-content-api-for-ai-agents for the API positioning and /docs for endpoint details.