本文へ移動
cccskills
無料GitHub で公開

snack-batching

Weekly snack-request batching for {{office_channel}}. Collects Slack shortcut submissions since the last order, consolidates duplicates and standing staples, builds a cart against {{preferred_vendor}} within {{weekly_budget}}, and posts it for the office manager to approve before anything is placed.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md5.5 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

<skill name="snack-batching"> <overview> Turn scattered snack requests into one approved order a week. A shortcut in {{office_channel}} is the one place people ask; a weekly cron spawns a fresh session that reads everything submitted since the last batch, consolidates it with the standing rules below, prepares a cart against {{preferred_vendor}}, and posts it for the office manager to approve. Nothing is placed until that approval lands back in the same channel.

Each run is stateless — there is no ledger. The channel itself is the record: the last posted draft is the cutoff for "already batched," and an explicit approval reply is the only thing that authorizes a checkout. </overview>

<when-to-load> - The weekly cadence cron fires the batching run. - Someone replies to a posted draft order in {{office_channel}} — to approve it, ask about it, or change a line item. - A human asks the agent to batch requests early or check the draft status. </when-to-load> <workflow>

Step 1 — Find this week's window

Search {{office_channel}} for the agent's own last "Draft snack order" post.

  • Found one → its timestamp is the cutoff; everything after it is unbatched.
  • None found (first run) → use the last 7 days as the cutoff.
  • Found one that's still awaiting approval → do not start a new batch. Go to Step 6 and check whether it's been approved instead.

Step 2 — Collect requests since the cutoff

Read every message in {{office_channel}} posted by the snack-request shortcut since the cutoff. Each submission has the shape:

New snack request: <item> — qty <n> — requested by <@user>

Parse item, qty, and requester out of each one. Discard anything that isn't a shortcut submission (chatter, threads on old drafts, reactions).

Step 3 — Consolidate and apply standing rules

  • Merge identical or near-identical items (e.g. "Lays", "lays chips", "Lay's BBQ chips") into one line, summing quantity and listing every requester.
  • Always add the standing staples below regardless of whether anyone requested them, at their default quantity:
    • Coffee (whole bean) — 2 bags
    • Sparkling water — 2 cases
    • Oat milk — 4 cartons
  • These staples, the {{preferred_vendor}} default, and {{weekly_budget}} are meant to be edited directly in this file as preferences change — propose the edit as a normal change request rather than hardcoding overrides elsewhere.

Step 4 — Build the cart, never check out

Using ORDERING_ACCOUNT_API_KEY, search {{preferred_vendor}} for each consolidated line item and add it to a cart (do not submit/checkout).

  • Running total over {{weekly_budget}}? Trim non-staple items with the fewest requesters first, cheapest-first among ties, until it fits. Never trim a standing staple to make budget.
  • Item unavailable? Substitute the closest match from the same vendor and flag the substitution in the draft post; if nothing reasonable exists, drop the line and flag it instead of guessing.

Step 5 — Post the draft for approval

Post one message to {{office_channel}}:

🧺 Draft snack order — week of <date range>

- <item> × <qty> — requested by <@user, @user2>
- ...
- Coffee (whole bean) × 2 — standing staple
- ...

Total: $<total> (budget: {{weekly_budget}})
Vendor: {{preferred_vendor}}

Nothing has been purchased. Reply "approved" in this thread to place this
order.

This is the only output of the run. Stop here — do not poll or wait inside the session.

Step 6 — Place the order only after explicit approval

When a reply in-thread on a draft is an explicit approval (e.g. "approved", "yes, place it", a clear thumbs-up plus a written yes from the office manager) — and only then:

  1. Check out the exact cart from Step 4 against {{preferred_vendor}} using ORDERING_ACCOUNT_API_KEY.
  2. Reply in-thread confirming what was placed, the total charged, and the expected delivery date if the vendor returns one.

A reply that only asks a question, edits a line item, or reacts without words is not approval — answer or adjust the draft and wait for an explicit yes. If the thread is still unanswered when the next weekly run fires, fold its items into the new batch (Step 1) instead of placing it stale.

</workflow> <guardrails> - **No checkout without an explicit, in-thread, written approval.** Silence, emoji alone, and side conversation never count. - **Checkout is the only irreversible step.** Everything before it (search, cart-build, draft post) is reversible and safe to redo. - **Scoped secret.** `ORDERING_ACCOUNT_API_KEY` is injected into the sandbox at runtime, scoped to this agent's grant. Never echo it into a reply, a log, or the Slack post. - **Isolation.** Each weekly run executes in its own sandbox; only the draft post and (post-approval) the checkout call leave it. - **One draft per week.** Don't post a second draft while one is still awaiting approval — resolve or roll it forward first (Step 1). - **Everything is code.** The vendor, the budget, and the standing staples live in this file, versioned and changed only through a reviewed change request — never as an ad hoc override during a run. </guardrails> </skill>

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Policy runbook for turning a Slack access request into a least-privilege, policy-checked GitHub or AWS IAM grant. Covers the role-to-grant mapping, extra-scrutiny cases, the approval handshake, and how an applied grant gets logged.

日本語の概要は準備中です。原文の説明を表示しています。

kortix-ai/suna2万2026年10月11日 更新

Get a complete picture of any company or person before outreach. This skill always works with web search, and gets significantly better with enrichment and CRM data.

日本語の概要は準備中です。原文の説明を表示しています。

kortix-ai/suna2万2026年10月11日 更新

Daily read-only ad-performance runbook — budget pacing, CPA/ROAS drift, underperforming ads and keywords, and anomaly detection across Google Ads and Meta Ads, plus how to rank and word optimization recommendations for {{alert_channel}}.

日本語の概要は準備中です。原文の説明を表示しています。

kortix-ai/suna2万2026年10月11日 更新

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.

日本語の概要は準備中です。原文の説明を表示しています。

kortix-ai/suna2万2026年10月11日 更新

Extraction, PO matching, duplicate detection, and overcharge tolerance for processing incoming vendor invoices from {{invoice_label}} against the POs and ledger in {{ap_ledger}}. Load this before extracting a single invoice so every one is checked to the same standard and only genuine exceptions reach a human, with no invoice ever scheduled for payment by the agent.

日本語の概要は準備中です。原文の説明を表示しています。

kortix-ai/suna2万2026年10月11日 更新

Daily brand-mention monitoring loop for {{brand_terms}}. Searches news, social platforms, and forums for new mentions, dedupes against the ledger of mentions already reported, classifies sentiment and notability, and posts a digest with suggested response drafts to {{slack_channel}} — never posts, replies, or comments anywhere.

日本語の概要は準備中です。原文の説明を表示しています。

kortix-ai/suna2万2026年10月11日 更新

kortix-ai のスキルをすべて見る

このスキルの問題を報告する