Daily engineering digest

What you tell the assistant

Every weekday at 9am Toronto time, post a digest to #engineering with open PRs waiting on review and bugs opened in the last day in northwind/api.

Why it works

A sandboxed JavaScript step groups and counts results — no model tokens spent on arithmetic.
Schedules run in your timezone.
The fast model writes the summary, keeping the run cheap.

Build this workflow

What it builds

Trigger

Schedule

0 9 * * 1-5 · America/Toronto

tool

Open pull requests

github.list_pull_requests

tool

New bugs

github.search_issues

javascript

Group and count

→ waiting, drafts, bug_count

llm

Write the digest

fast

tool

Post to #engineering

slack.post_message

View the definition

trigger:
  kind: schedule
  cron: "0 9 * * 1-5"
  timezone: America/Toronto
steps:
  - key: prs
    name: Open pull requests
    kind: tool
    tool: github.list_pull_requests
    args: { repo: northwind/api, state: open, limit: 50 }
  - key: bugs
    name: New bugs
    kind: tool
    tool: github.search_issues
    args: { query: "repo:northwind/api is:issue is:open label:bug created:>={{ run.started_at }}", limit: 50 }
  - key: shape
    name: Group and count
    kind: javascript
    inputs: { prs: "{{ steps.prs.output }}", bugs: "{{ steps.bugs.output }}" }
    outputs: { waiting: array, drafts: integer, bug_count: integer }
    code: |
      function main({ prs, bugs }) {
        const waiting = prs.filter(pr => !pr.draft).map(pr => `#${pr.number} ${pr.title} (@${pr.author})`)
        console.log(`${waiting.length} ready for review, ${bugs.length} new bugs`)
        return { waiting, drafts: prs.length - waiting.length, bug_count: bugs.length }
      }
  - key: write
    name: Write the digest
    kind: llm
    model: fast
    prompt: |
      Write a short, friendly Slack digest for the engineering team.
      Ready for review: {{ steps.shape.output.waiting }}
      Drafts: {{ steps.shape.output.drafts }}. New bugs: {{ steps.bugs.output }}
  - key: post
    name: "Post to #engineering"
    kind: tool
    tool: slack.post_message
    args: { channel: C01ENGINEER, text: "{{ steps.write.output.text }}" }

A sample run

What each step produces

Open pull requests tool

succeeded

Calling github.list_pull_requests

[
  {
    "number": 1482,
    "title": "Backfill invoice currency",
    "author": "sokafor",
    "draft": false
  },
  {
    "number": 1479,
    "title": "Cache plan lookups",
    "author": "pramen",
    "draft": false
  },
  {
    "number": 1476,
    "title": "Experiment: new pricing page",
    "author": "lmartins",
    "draft": true
  }
]

New bugs tool

succeeded

Calling github.search_issues

[
  {
    "number": 1485,
    "title": "CSV export times out for large accounts",
    "author": "mchen"
  },
  {
    "number": 1484,
    "title": "Webhook retries duplicate invoices",
    "author": "sokafor"
  }
]

Group and count javascript

succeeded

2 ready for review, 2 new bugs

{
  "waiting": [
    "#1482 Backfill invoice currency (@sokafor)",
    "#1479 Cache plan lookups (@pramen)"
  ],
  "drafts": 1,
  "bug_count": 2
}

Write the digest llm

succeeded

Using model fast (Claude Haiku 4.5, Murmurator AI)

{
  "text": "Good morning! :sunny: *2 PRs are ready for review*: #1482 Backfill invoice currency, #1479 Cache plan lookups. 1 draft in progress. *2 new bugs* came in — the CSV export timeout looks customer-facing."
}

Post to #engineering tool

succeeded

Calling slack.post_message

{
  "channel": "C01ENGINEER",
  "ts": "1758632400.000200"
}

Trigger payload

{
  "scheduled_at": "2026-09-23T13:00:00Z"
}

More examples

Trigger

GitHub

github · pull_request · northwind/api

tool

Fetch pull request

github.get_pull_request

github.create_comment

tool

Alert #eng-reviews

slack.post_message

if review.output.data.risk equals high

GitHubSlackYour AI models

PR review bot

A careful first review on every pull request, in under a minute.

Trigger

GitHub

github · issues · northwind/api

tool

Read the issue

github.get_issue

if trigger.label.name equals automate

tool

Clone into a workspace

workspace.create

tool

Hand it to the coding agent

if tests.output.exit_code equals 0

tool

Report a failing suite

github.create_comment

if tests.output.exit_code not equals 0

Open the pull request

github.create_pull_request

GitHubYour AI models

Issue to pull request

Label an issue `automate`, get a pull request with passing tests.

Research the question

SlackPostgreSQLYour AI models

Ask your warehouse in Slack

Mention the bot, get an answer backed by a real query.

Your next automation is one sentence away.

14-day free trial with $5 of built-in AI included. Cancel anytime.

Start free trial Talk to us