Weekly Code Health Report for Engineering Leaders

PromptCreate

Every Monday at 8am, write a code health brief for our engineering leadership so nobody has to open a SonarCloud dashboard to know how we are doing.

Start with the SonarCloud "Projects — Search" operation to enumerate every project in our organization. Page through the results so the brief covers all of them, not just the first page.

For each project, use "Measures — Component" to read the current coverage, duplicated lines density, technical debt (the sqale_index metric), and the reliability, security and maintainability ratings. Then use "Quality Gates — Project Status" to get the current pass or fail state.

Use "Measures — Search History" to pull those same metrics from roughly one week ago for each project, and compare. This comparison is the whole point of the brief: lead with what actually moved rather than a wall of raw numbers.

In the narrative, call out the three projects that regressed the most and explicitly flag any project whose coverage fell by more than two percentage points. Also name anything that newly failed its quality gate, and give credit where a project genuinely improved.

If a project has no fresh analysis in the past week, or is missing one of the metrics, leave it out of the comparison rather than reporting a false regression, and note at the end of the brief which projects had no new data.

Append one row per project per week to a tracking spreadsheet using the Google Sheets "Append Values" operation so a trend line builds over time. Each row should carry the week's date, the project name, coverage, duplicated lines density, technical debt, the three ratings, and the quality gate state.

Then send the narrative brief to the engineering leads with the Gmail "Send a Message" operation. Keep the prose skimmable for a VP who does not read code: short paragraphs, plain language, no metric jargon, and a one line verdict at the top saying whether code health improved or declined overall this week. Let the spreadsheet carry the per project detail, and link to it from the email.

Example output

Code health, week of 9 June Overall: slightly down. Two projects slipped, one recovered, and the rest held steady. What moved most: 1. checkout-service. Coverage fell from 71% to 64% and it is now failing its quality gate. This is the one to ask about. 2. billing-api. Technical debt grew by about 4 days of work, mostly new maintainability issues from last week's refactor. 3. notifications-worker. Duplicated code rose from 3.1% to 5.8%. Coverage alert: checkout-service dropped 7 points, well past the 2 point threshold. Good news: internal-admin cleared its quality gate for the first time in a month. Full per project numbers are in the tracking sheet.

What does this prompt do?

Reads every project in your SonarCloud organization each Monday morning and pulls the numbers that matter: test coverage, duplicated code, technical debt, and the reliability, security and maintainability ratings.
Compares this week against last week and leads with what actually changed, naming the three projects that slipped the most.
Flags any project whose test coverage dropped by more than two points, and any project that is now failing its quality gate.
Logs one row per project per week to a spreadsheet so a trend builds over months, then emails the written brief to your engineering leads.

What do I need to use this?

A SonarCloud account covering the projects you want in the brief
A Google account for Gmail and Google Sheets
A spreadsheet to hold the weekly tracking rows
The email addresses of the engineering leads who should receive the brief

How can I customize it?

Change the timing: Monday at 8am is the default, but any day and hour works.
Change who gets it: send to a single VP, a leads alias, or a wider engineering list.
Change the thresholds: the two point coverage drop and the top three regressions are both easy to adjust.
Narrow the scope to a subset of projects if you only care about what is running in production.

FAQs

Do I need to understand the SonarCloud dashboard to read this?

No, and that is the point. The email is written in plain language for someone who does not read code. The detailed numbers live in the spreadsheet for anyone who wants to dig in.

What is technical debt in this report?

It is your code quality tool's estimate of how much engineering time it would take to fix the known issues in a project. Tracking it week over week shows whether that backlog is growing or shrinking.

Will it still email us on a quiet week?

Yes, you still get the Monday brief, but it will say plainly that little changed rather than padding it out with numbers. You can also adjust it to stay quiet when nothing meaningful moved.

Does this work if we only have a handful of projects?

Yes. The brief covers however many projects exist in your organization, whether that is three or three hundred.

Can the brief go somewhere other than email?

Yes. The brief is just written text, so it can be pointed at a team chat channel instead if that is where your leads actually read things.

Related templates

Trace phishing emails to the sending IP and report abuse

Every 15 minutes, forwarded phishing reports get traced back to the server that really sent them, with a verdict in Slack and the worst senders reported.

Agentic Task

Weekly Amazon S3 bucket security audit posted to Slack

Every Monday, check every S3 bucket for public exposure, missing encryption and weak backup settings, then get the risks ranked in Slack.

Agentic Task

Turn procurement portal tenders into CRM deals each morning

Every weekday at 7am, sign in to the tender portals you track, filter new notices against your bid criteria, and open a deal for the ones worth chasing.