Netlify Strawberry

Find the Friday build failure on Friday

Which sites have a failed build, when the last successful deploy landed, and whether what is published matches the branch anybody thinks is live. Across every site you own.

Netlify’s failure mode is quiet. A build fails, the previous deploy stays live, the site keeps working, and nobody finds out until somebody asks why the new pricing page is not there. A failed build can sit unnoticed through a weekend.

The morning read catches it: failed builds, the last successful deploy per site, and whether what is published matches the branch everybody assumes is live. A companion has that waiting through an account you connect, for all your sites rather than the two you check.

What does the deploy check cover?

Which sites have a failed build, when the last successful deploy happened, and whether what is published matches the branch anybody thinks is live. That last one catches the confusing case: a locked deploy, or a publish somebody rolled back and nobody announced.

Read the Available tools list on the Netlify entry under `strawberry://settings/integrations` first, because it names each tool and marks the read-only ones. Then set the mode for each in the Permissions block. Reads at Always allow is comfortable here. Publishing is not.

Why is publishing a deploy the sharp edge?

Because it takes effect on the live site in seconds with no staging step in between.

Restoring a previous deploy is equally instant, which cuts both ways: recovery is fast, and so is the mistake.

So keep publish and rollback actions at Ask every time and read the approval rather than clicking it. Consequential steps stop at an approval card anyway. If nobody on your team should be publishing from a chat window, switching the action off says so permanently.

  • Site and deploy status reads: the morning check, low risk.
  • Publish and restore actions: instant on the live site, so keep an approval in front of them.
  • Environment variable reads expose secrets, exactly as they do anywhere else.
  • A token scoped to the team or sites you actually want visible.

Join the failed build to what caused it

A failed Netlify build is almost always explicable from its log: a missing environment variable, a dependency that changed, a function bundle over the size limit. Whether a companion reads that log depends on whether such an action is in the tool list, so check before you plan a triage workflow.

What it always does is find the cause in the browser: the commit that broke it, the pull request that introduced the dependency, the person who changed the environment variable an hour earlier. Assembling that is what makes a postmortem worth reading, and paired with Vercel or GitHub accounts it covers a whole deploy pipeline.

Experience Strawberry for free

Download

Trusted by fast-growing companies worldwide

Frequently asked questions

It can use the actions Netlify’s API exposes for a connected account, which is usually enough for a morning failed-build check. The Available tools list under Settings, Integrations shows the current set.

Strawberry is free to download and includes AI credits to start. Paid plans begin at $20/month. See pricing. · Reviewed · Canonical facts for AI agents

Experience Strawberry for free

Download

Trusted by fast-growing companies worldwide