> For the complete documentation index, see [llms.txt](https://docs.shortcutco.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.shortcutco.com/developers/guides/daily-schedule-bot.md). # Daily schedule bot Post each crew's route for the day to Slack every morning with a short script and a read-only token. After this guide, a message like this lands in your team's Slack channel every morning before the crews roll out: ``` Schedule for Friday, October 16 (14 visits) North crew (8) 1. Robin Alvarez, 412 Elm St, Bozeman, MT 59715 (Weekly mowing) 2. Okafor Dental, 1180 N 19th Ave, Bozeman, MT 59718 (Weekly mowing) ... South crew (5) ... Unassigned (1) - Jamie Chen, 88 Willow Way, Bozeman, MT 59718 (Aeration) ``` It takes about 15 minutes: a read-only token, a Slack incoming webhook, and one script you run on a schedule. No AI is involved; the script calls the Shortcut MCP server's `get_schedule` tool directly over HTTPS, because that tool returns the day already grouped by crew in route order, with crew and customer names, in one call. ## 1. Create a read-only token In **Settings → API & integrations**, select **Create token**: * **Name:** `Schedule bot`. * **Scopes:** **Custom** with **Read visits**, **Read crews**, and **Read customers**. Read crews adds crew names and Read customers adds customer names (the company name for businesses); the schedule itself needs only Read visits. * **Expiry:** 1 year. ## 2. Create a Slack incoming webhook In Slack, create an app with **Incoming Webhooks** turned on for the channel the crews read, and copy the webhook URL. Slack's guide: [Sending messages using incoming webhooks](https://api.slack.com/messaging/webhooks). Any chat tool that accepts a JSON `POST` works the same way. ## 3. Try the call ```bash curl https://api.shortcutco.com/mcp \ -X POST \ -H "Authorization: Bearer $SHORTCUT_TOKEN" \ -H "Content-Type: application/json" \ -H "Accept: application/json, text/event-stream" \ -H "MCP-Protocol-Version: 2025-11-25" \ -d '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "get_schedule", "arguments": {}}}' ``` With no arguments, `get_schedule` returns today in your workspace's timezone. The answer's `result.structuredContent` holds the schedule (the same JSON is in `result.content[0].text` as a string, left out here): ```json { "jsonrpc": "2.0", "id": 1, "result": { "content": [{ "type": "text", "text": "…" }], "structuredContent": { "data": { "today": "2026-10-16", "from": "2026-10-16", "to": "2026-10-16", "totalVisits": 2, "days": [ { "date": "2026-10-16", "totalVisits": 2, "teams": [ { "teamId": "6c7d8e9f-0a1b-4c2d-8e3f-4a5b6c7d8e9f", "totalVisits": 1, "visits": [ { "id": "5e0c8a1f-2b7d-4c3e-9f6a-8d1b2c3e4f50", "object": "visit", "date": "2026-10-16", "status": "scheduled", "closeReason": null, "jobId": "a4f1c9e2-7b3d-4e8a-9c5f-1d2e3f4a5b6c", "jobName": "Weekly mowing", "customerId": "9a8b7c6d-5e4f-4a3b-8c2d-1e0f9a8b7c6d", "propertyId": "1b2c3d4e-5f6a-4b7c-8d9e-0f1a2b3c4d5e", "address": "412 Elm St, Bozeman, MT 59715", "assignedTeamId": null, "effectiveTeamId": "6c7d8e9f-0a1b-4c2d-8e3f-4a5b6c7d8e9f", "routePosition": 1, "plannedStartAt": "2026-10-16T14:00:00.000Z", "plannedEndAt": "2026-10-16T14:30:00.000Z", "estimatedDurationMinutes": 30, "arrivedAt": null, "completedAt": null, "invoiceId": null, "createdAt": "2026-04-02T15:12:44.120Z", "updatedAt": "2026-10-14T21:03:10.552Z" } ] }, { "teamId": null, "totalVisits": 1, "visits": [ { "id": "8b9c0d1e-2f3a-4b4c-9d5e-6f7a8b9c0d1e", "object": "visit", "date": "2026-10-16", "status": "scheduled", "closeReason": null, "jobId": "3c4d5e6f-7a8b-4c9d-8e0f-1a2b3c4d5e6f", "jobName": "Aeration", "customerId": "2a3b4c5d-6e7f-4a8b-9c0d-1e2f3a4b5c6d", "propertyId": "7f8e9d0c-1b2a-4c3d-8e4f-5a6b7c8d9e0f", "address": "88 Willow Way, Bozeman, MT 59718", "assignedTeamId": null, "effectiveTeamId": null, "routePosition": null, "plannedStartAt": null, "plannedEndAt": null, "estimatedDurationMinutes": 45, "arrivedAt": null, "completedAt": null, "invoiceId": null, "createdAt": "2026-10-16T19:40:02.511Z", "updatedAt": "2026-10-16T19:40:02.511Z" } ] } ] } ], "teamNames": { "6c7d8e9f-0a1b-4c2d-8e3f-4a5b6c7d8e9f": "North crew" }, "customerNames": { "9a8b7c6d-5e4f-4a3b-8c2d-1e0f9a8b7c6d": "Robin Alvarez", "2a3b4c5d-6e7f-4a8b-9c0d-1e2f3a4b5c6d": "Jamie Chen" } }, "meta": { "hasMore": false, "nextCursor": null } } } } ``` Visits come in schedule order: by date, then crew (unassigned last, with `teamId: null`), then route position. A page holds at most 50 visits; on a bigger day `meta.hasMore` is `true` and you pass `meta.nextCursor` back as the `cursor` argument. `totalVisits` on each crew tells you when a crew's list continues on the next page. ## 4. The script ```js // schedule-bot.mjs: Node 18 or later, no dependencies. // SHORTCUT_TOKEN=scut_live_... SLACK_WEBHOOK_URL=https://hooks.slack.com/... node schedule-bot.mjs async function getSchedulePage(cursor) { const response = await fetch("https://api.shortcutco.com/mcp", { method: "POST", headers: { Authorization: `Bearer ${process.env.SHORTCUT_TOKEN}`, "Content-Type": "application/json", Accept: "application/json, text/event-stream", "MCP-Protocol-Version": "2025-11-25", }, body: JSON.stringify({ jsonrpc: "2.0", id: 1, method: "tools/call", params: { name: "get_schedule", arguments: cursor ? { cursor } : {} }, }), }); const message = await response.json(); const requestId = response.headers.get("x-request-id"); if (message.error) throw new Error(`${message.error.message} (request ${requestId})`); if (message.result.isError) { const { code, message: text } = message.result.structuredContent.error; throw new Error(`${code}: ${text} (request ${requestId})`); } return message.result.structuredContent; } // Read every page, merging crews that continue across pages. const crews = new Map(); const names = { teams: {}, customers: {} }; let today; let cursor = null; do { const { data, meta } = await getSchedulePage(cursor); today = data.today; Object.assign(names.teams, data.teamNames); Object.assign(names.customers, data.customerNames); for (const day of data.days) { for (const team of day.teams) { const key = team.teamId ?? "unassigned"; if (!crews.has(key)) crews.set(key, { teamId: team.teamId, visits: [] }); crews.get(key).visits.push(...team.visits); } } cursor = meta.nextCursor; } while (cursor); const heading = new Date(`${today}T12:00:00Z`).toLocaleDateString("en-US", { weekday: "long", month: "long", day: "numeric", timeZone: "UTC", }); const total = [...crews.values()].reduce((sum, crew) => sum + crew.visits.length, 0); const lines = [`*Schedule for ${heading}* (${total} visits)`]; for (const crew of crews.values()) { const crewName = crew.teamId ? names.teams[crew.teamId] ?? "Crew" : "Unassigned"; lines.push("", `*${crewName}* (${crew.visits.length})`); for (const visit of crew.visits) { const stop = visit.routePosition ? `${visit.routePosition}.` : "-"; const customer = names.customers[visit.customerId] ?? "Customer"; lines.push(` ${stop} ${customer}, ${visit.address ?? "no address"} (${visit.jobName})`); } } if (total === 0) lines.push("", "Nothing on the schedule today."); const posted = await fetch(process.env.SLACK_WEBHOOK_URL, { method: "POST", headers: { "Content-Type": "application/json" }, body: JSON.stringify({ text: lines.join("\n") }), }); if (!posted.ok) throw new Error(`Slack answered ${posted.status}`); ``` Run it once by hand and check the channel: ```bash SHORTCUT_TOKEN="scut_live_paste_your_token_here" SLACK_WEBHOOK_URL="https://hooks.slack.com/services/paste/your/url" node schedule-bot.mjs ``` The heading uses `today` from the answer, the date in your workspace's timezone, so the bot is right wherever it runs. ## 5. Run it every morning Any scheduler works. With cron on a server whose clock is on your business's time, every day at 6:00 AM: ``` 0 6 * * * cd /opt/schedule-bot && SHORTCUT_TOKEN=scut_live_... SLACK_WEBHOOK_URL=https://hooks.slack.com/... node schedule-bot.mjs ``` On GitHub Actions, a scheduled workflow runs it with the token and webhook URL stored as repository secrets (`cron` there is in UTC: 6:00 AM in Denver is `0 12 * * *` in summer and `0 13 * * *` in winter). ## Variations * **One crew per channel.** Pass `{ teamId: "…" }` as the `get_schedule` arguments to get one crew's visits, and post each crew to its own channel. * **Tomorrow, the evening before.** Pass `{ from: "2026-10-17" }`; compute tomorrow from `today` in the first answer. * **The whole week.** Pass `from` and `to` up to 31 days apart; the answer groups by date first. * **Straight from the REST API.** `GET /v1/visits?from=2026-10-16&to=2026-10-16` returns the day's visits, including `pending` ones that are not on the schedule yet, newest first, each with `effectiveTeamId` and `routePosition`. Filter by `status`, sort by crew and route position, and take crew names from `GET /v1/teams` to build the same list.