Prepare meetings and follow-ups from local work memory with Rowboat
Use Rowboat to build local work memory from email, calendar, notes, and web context, then generate meeting briefs, follow-ups, docs, and decks.
npx skills add agentskillexchange/skills --skill prepare-meetings-and-follow-ups-from-local-work-memory-with-rowboat
Use Rowboat when an agent needs durable work context instead of rebuilding history from scattered email, meetings, notes, and documents each time. The operator connects Gmail, Google Calendar, meeting notes, local Markdown vaults, optional web search, and MCP or Composio tools; Rowboat turns that material into an inspectable local knowledge graph and uses it to prepare meeting briefs, draft grounded follow-ups, track people or topics, and produce docs or PDF slides. Inputs are connected accounts, notes, vault content, topic/person trackers, API keys for optional tools, and the user’s current task. Outputs are Markdown-backed memory updates, briefs, follow-up drafts, live notes, documents, decks, and reviewable local actions. The scope boundary is personal/work memory and context-grounded operator output, not a generic AI assistant, note-taking app, email client, or product listing.
What this skill actually does
Inputs and prerequisites: Rowboat desktop app, connected email/calendar or notes sources, local Markdown vault, optional MCP or Composio tools, and optional Deepgram, ElevenLabs, or Exa API keys.
Setup notes: Download the latest Rowboat app for Mac, Windows, or Linux from https://www.rowboatlabs.com/downloads or the GitHub releases page, complete Google setup for Gmail, Calendar, and Drive if needed, then connect notes, meeting sources, optional MCP tools, and model/API providers before running meeting-prep, follow-up, live-note, or document-generation workflows.
Source and verification boundary: use https://www.rowboatlabs.com as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.