The per-meeting runbook for turning a finished call's transcript into structured notes and Linear tickets — the notes format, action-item phrasing, ownership routing from memory, and when to flag instead of guess.
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。
Pull the calendar event (title, attendees, agenda) via google_calendar. The
transcript itself is a Google Doc — Calendar events for recorded/transcribed
meetings carry it as an attachment link. Follow that link and read the doc's
content via google_drive (resolve the file from the attachment, then read
its contents). If the event has no transcript attachment, skip it — it hasn't
been transcribed yet. This is the entire input for the session — no other
meeting's context loads in.
Read the notes template, the phrasing convention for an action item, and the ownership map (who owns which area or project) from memory. Skim recent past meetings on the same topic for continuity — a decision this call revisits should reference the earlier one, not restate it as new.
Write the notes in the standard three-part shape:
| Section | Content |
|---|---|
| Decisions | What was agreed, stated as a fact ("Ship Friday"), not a discussion point |
| Discussion | The context and trade-offs behind each decision, condensed |
| Next steps | Every action item, phrased "[Owner] will [do X] by [when]" |
Attach the notes to the meeting record so they live where the team already looks for them.
| Signal | Routing |
|---|---|
| Owner named explicitly in the transcript ("Sarah will handle the migration") | Assign to that person |
| Owner implied by the ownership map (topic maps to a known area owner) | Assign to the mapped owner |
| No explicit owner and no confident map match | Do not guess — flag it (Step 6) |
Create one ticket per action item in {{linear_team}}: title is the action phrased as a task, description is the decision/discussion context plus a link back to the meeting record, assignee is the owner resolved in Step 4.
File any action item without a confident owner as an unassigned ticket in {{linear_team}}, with the ambiguity spelled out in the description ("raised by X, no clear owner — needs triage") instead of guessing an assignee.
Once the notes are attached and every action item is either filed-and-assigned or filed-and-flagged, the session is done. Don't touch any ticket that already existed before this meeting.
</workflow> <guardrails> - **Only create.** The agent never closes, reassigns, or edits a ticket that existed before this session — Linear writes are new tickets only. - **Flag instead of guess.** An action item without a confident owner is filed unassigned with the ambiguity noted, never routed to a person on a guess. - **One meeting, one session.** Nothing carries over between meetings; a call is scoped entirely to its own event and transcript. - **Scoped secrets.** Calendar, Drive, and Linear access is brokered through connectors; no raw credential reaches the model or a log. - **No chat posts.** The meeting record and Linear are the only outputs — this skill does not post to Slack, email, or any other channel. </guardrails> </skill>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。
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。
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。
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。
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。
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。
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。