Social account connection health console for client brands

Build me a connection health console for the client brands I manage in bundle.social. I run social accounts for a lot of clients and I need one screen I can open before I trust anything to publish. The subject of this app is connection state, quota headroom and orphaned scheduled posts across every client. It is deliberately preventative and cross-client, so keep it centered on the health grid and do not turn it into a failure alert surface.

The main view is a grid. Down the side is every bundle.social team, loaded with the Get list of teams operation. Across the top is every connected platform. Build each row with the Get team operation, which returns that team's connected social accounts and settings, and use Get social account by team and type when you need the detail behind a single cell. Color-code each cell by whether that account is still authorized, whether it needs reconnecting, or whether that platform was never connected for that client. Resources nest as organization then team: one API key authenticates the organization and teamId scopes each request, so a single credential genuinely can render an all-clients grid.

Above the grid, show platform status from the Get health operation, which reports API reachability plus the current operational status of every supported platform. If Meta is having an incident, mark the whole Instagram and Facebook columns as a platform issue so I can see the red is not my client's fault. State in the interface that Meta access tokens are commonly cited at roughly a 60 day lifespan, so Instagram and Facebook are the columns that need the most attention.

Also above the grid, add a plan capacity strip fed by the Get organization posts usage, Get organization comments usage and Get organization uploads usage operations. Show consumed against the plan limit for the current billing period and flag anything close to its cap, warning around 80 percent and critical around 95 percent.

Selecting a cell opens a detail panel and runs the Manually run connection/disconnect check operation on the spot, so I get a live answer rather than a cached one. If the authorization is dead, let me generate a hosted reconnect link with the Create portal link operation, which runs the whole OAuth and channel picker flow, and send it to that client's channel with the Slack Bot Send a Message operation instead of writing the same reconnect email by hand. Let me map each team to a Slack channel using the Slack Bot List Channels operation, and persist that mapping so I only set it once.

Because disconnecting an account also removes it from all scheduled posts, add an orphaned posts view. List the posts stranded by a disconnect with the Get posts that can be reconnected to a social account operation, and let me reattach them in one click with the Reconnect social account to matching posts operation, so scheduled content does not just quietly fail to go out.

Let me sort clients by risk. The ones that should sit on top are those with an expired connection and posts scheduled in the next 48 hours. Use the Get post list operation to find scheduled posts in that window, and use it again filtered to status ERROR to fill a symptom column of recent publishing failures per client. Publishing failures are reported per platform under errorsVerbose with code, userFacingMessage and isTransient, so surface userFacingMessage and use isTransient to tell a dead token apart from a transient blip.

Persist the per-team Slack channel mapping and the timestamp of the last manual connection check for each account, so the grid can show how fresh each cell is and I can tell a stale green from a verified one.