Automate a weekly GitHub stars and activity report for your repos

Maintainers want to know which repositories are growing but notice a spike by chance. A scheduled agent reads public star counts weekly, compares them with last week's report, and drafts a short report you approve.

What you’ll set up

Connections

Works from public sources: GitHub REST API: public repository star counts and activity

Upgrades

Delivery (channel or inbox)

Delivers the approved output to your channel or inbox.

Inputs

Repositories (owner/name) · Required

List, no default

Launch or post notes · Optional

Text, no default

Ready by · Required

Time, default 09:00

Brief · Required

Text, prefilled and editable during setup.

Runs

Runs every Monday at 08:00

Done by Market researcher

You approve before anything goes out.

Set up this job

You approve every report · Free to start

How it runs

One agent does the work. Its department head reviews it, then you approve.

  1. 1Market researcher

    Runs every Monday at 08:00 and prepares the report from your connected sources.

  2. 2CMO · Review

    The department head reviews the report before it reaches you.

  3. 3You

    You approve before anything goes out.

What you get

Sample, not live dataSample stars report (Sample, not live data)

Sample, not live data. Repositories and numbers are invented.

Stars, week of 3 Nov (sample)

example/cli-tool 1,204 stars (+38). 1 new release, 14 commits, 9 open issues.

example/docs-site 311 stars (+2). No release, 3 commits, 1 open issue.

Note: +38 follows the launch post on 29 Oct (added by owner).

Needs your approval before it is sent.

What stars can and cannot tell you

Stars are a rough signal of attention, not usage. Still, a jump after a post or a launch tells you which channel worked, and a quiet stretch tells you the project needs a push. The count on a repository page is a snapshot; to see change you need last week's number, which nobody writes down. The job is to write it down for you and say what moved.

How the scheduled job works

You list the public repositories to track. Each Monday a scheduled agent reads each repository's star count and open issue count through the GitHub repository endpoint, counts commits since the previous report with the list commits endpoint, and checks for a new release. It compares the numbers with last week's report and shows the change in stars, new releases, commits and open issues. If last week's report is not available, it records a baseline, so growth appears from the second week. It reads public data and writes nothing to GitHub.

Review and approval

The CMO agent, as department head, reviews the report before it reaches you: do the changes match last week's figures, is a spike explained by a known launch or post, is anything compared against the wrong week. Then you approve, edit or skip it, and nothing is sent until you do. The report does not say who starred or why, so it will not guess at causes; add a note when you know of one.

Using the report

Pair a spike with what you published that week and write it in the launch or post notes field. After a month you will see which posts, directories or mentions moved the number and which did nothing. Keep the list short, a handful of repositories, so each line gets read. A repository that never moves is useful information too: it tells you where promotion has not landed yet.

Questions

Start with one report

Set up this job

You approve every report · Free to start