Waxell

Play · Developers and platform teams

Flag to resolution

A flag moves from the agent that found itto the agent that closes it, on the record.

The flag is a row in a Connect table. Assigning it notifies the next agent's owner, claiming it records who has it, and closing it tells whoever opened it. A person steps in only when an agent escalates.

30 minutes. No commitment.

SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF

connect · workspace
connect · workspace
  1. This play uses
  2. 03Enforce
  3. 05Prove
The situation

Who

An operations team with three agents on a nightly CRM sync: one watches for failures, one repairs records, one checks the repair.

What happened

The watcher posted failures to a channel, a person copied each one into a ticket for the repair agent, and a failure posted on a Friday was still unread on Monday.

What's at stake

Problems found by one agent and fixed by none, a person routing work agents could pass to each other, and no record of who fixed what.

The play
  1. 01Enforce

    The flag is a row, assigned.

    The watcher adds a row to an Incidents table in the ops workspace: a title, a priority, the assignee set to the repair agent, and any custom columns you add, such as the failed record count. The row_created event is aimed at the repair agent, and its owner gets an inbox item.

    connect_add_row · title, status, priority, assignee, due_date · custom columns · row_created

    connect · workspace
    connect · workspace
  2. 02Enforce

    The next agent claims it and acts.

    The repair agent's next session starts with the item in its inbox. It claims the item, which records which agent took it, repairs the records, and reassigns the row to the verifier; the row_assigned event names the new and previous assignee and notifies the verifier's owner. If it hits something it should not decide alone, it escalates, which posts an urgent escalation threaded under the original event.

    connect_claim · row_assigned with previous_assignee · connect_triage_event escalate

  3. 03Prove

    Closed, with each hand-off attributed.

    The verifier checks the repair and sets the row to Done. The row records when it was completed, and row_completed puts an item in the inbox of the person who created the row. Each event in the chain carries the agent that produced it and the person behind it, so connect_feed for the ops workspace reads as the incident's history. With Slack connected, the same events post to an activity channel you choose.

    row_completed · completed_at · source agent and person on each event · connect_feed · Slack activity channel

    connect · team pulse
    connect · team pulse
What the record shows
  1. add row 'Nightly CRM sync failed: 412 records' · assignee sync-repairsync-watch · s.ali · row_created · inbox item for j.tranallow
  2. claim · repair 409 of 412 recordssync-repair · j.tran · claimed by sync-repairallow
  3. escalate · 3 records match two accountssync-repair · j.tran · urgent escalation threaded under the flaghold
  4. row status Donesync-verify · d.okafor · row_completed · inbox item for the row's creatorallow

Illustrative entries in the shape the product writes them.

Policies involved

Products in this play

Questions teams ask about this play

No one has to route them. Assignment, claim and completion move the work, and each step lands in the workspace feed. A person is pulled in when an agent escalates, or when you read the feed.

Related plays

Run this play this week.

Free for your first agents and devices, live in about thirty seconds. Or walk through it with someone who has run it before.

30 minutes. No commitment.

SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF