Agent Ops Club by Reinventing.AI

Engineering and maintenance

Web Dev Employee

Keep website maintenance moving through reviewable changes.

Finds what broke on your site and fixes it for you to approve.

Free with an accountMIT licensed
  • 3 weekdays
  • 3 weekly
  • 2 monthly
Web Dev Employee with the tools of the role

One routine, start to finish

What lands when the Web Dev Employee runs.

01

Check every page.

A website creates a steady queue of maintenance: broken routes, production errors, dependency updates and small improvements. The Web Dev Employee checks the projects you name, groups errors into incidents, and works one ready fix at a time.

02

Fix on a branch. Test the fix.

Its output is a reviewable change with validation and a rollback path. Weekly maintenance and monthly inventory reviews keep the queue connected to the systems you actually run, while the report distinguishes a prepared fix from a change that has merged.

03

Hand over the change.

The shipped setup holds merges and deployments until you release them. Platform checks are read only, and a passing local check alone is not proof that the affected public page works.

A sample run, built from the kit's own routine. Nothing here is a customer's data.

Who is this for?

Three people the Web Dev Employee was built for.

  • Founder maintaining a shipped product

    Broken routes and errors found before a customer reports them.

  • Small team with recurring site issues

    Small fixes on a branch, tested on the page that broke.

  • Provider maintaining client sites

    A dated maintenance record for every site you look after.

The 8 scheduled routines.

The cadence is part of the role. Each one owns a defined piece of work.

  • Standup

    Resolves which drafted changes merged by reading the commit graph, closes and reopens incidents on that evidence, and writes the morning brief.

    Weekdays
  • Site sweep

    Checks every declared path from the outside, records one dated observation per project, then groups every new production error into a fingerprint.

    Weekdays
  • Fix runner

    Turns the card the standup marked next into a reviewable change on a branch, behind that project's own gate, with a rollback line attached.

    Weekdays
  • Platform guard

    Confirms the registrar, the host, and the hosted database still match the inventory, read only, and names every drift with both values.

    Weekly
  • Dependency run

    Works dependency and supply chain debt one project at a time, bumping the patch and lockfile class on a branch and holding the rest with evidence.

    Weekly
  • Weekly report

    Scores the week from the ledgers with a source path beside every number, and replays every learned browser flow once, read only.

    Weekly
  • Inventory refresh

    First run is the whole setup: it discovers what you actually run, writes the inventory and the opening policy, and registers the schedule.

    Monthly
  • Guardrail review

    Rewrites what the fix runner may change unsupervised, from a month of real merge outcomes. Widening waits for a streak; narrowing is immediate.

    Monthly
More about the Web Dev Employee: outputs, connections, setup and related material

A website creates a steady queue of maintenance: broken routes, production errors, dependency updates and small improvements. The Web Dev Employee checks the projects you name, groups errors into incidents, and works one ready fix at a time.

Its output is a reviewable change with validation and a rollback path. Weekly maintenance and monthly inventory reviews keep the queue connected to the systems you actually run, while the report distinguishes a prepared fix from a change that has merged.

Best fit: Founders maintaining shipped products, small teams with recurring site issues, and providers responsible for client website maintenance.

The work you can review

  • Dated site checks and grouped production incidents
  • Small fixes prepared on branches for review
  • Dependency and platform inventory reviews
  • Reports linking changes to validation and merge evidence

Set up the first week

Confirm the repositories, hosts and routes in scope. Review the first fix and its validation, define the protected branches, and keep merge and deploy permissions narrow while you establish the standard.

Choose the role in the Club install prompt builder, extract the kit to a local folder, and ask your agent to install it. The setup guide explains local and cloud harnesses. Your first week is supervision: inspect the output and correct the standard before widening permissions.

Accounts and connections

Logs, errors, advisors and the commit graph are read through your existing tools, and the render runs in a headless browser of its own. Connect only the services you use and the installed kit requires; access and costs depend on your accounts.