The Closer: a research brief before every Cal.com call

Use the Autoploy MCP for every read and save in this task. If it is not connected, stop and tell me instead of improvising. What to build: The Closer: a research brief before every booked call. When someone books a call, research who booked and write a short brief and talking points in your saved voice, posted to your team chat and logged in your Call briefs list. Runs on your Claude subscription, or on your own AI key. Autoploy never bills you for the AI. This is for founders and non-technical builders. How it behaves: 1. Runs when Cal.com posts a booking event to the automation. 2. Read who booked, their company, and the call time from the inbound booking. 3. If new booking, then go to "Check the Call briefs list", otherwise go to "Skip". 4. Look in the Call briefs list for a brief already written for this booking. 5. If already briefed this booking, then go to "Skip", otherwise go to "Read your business context". 6. A brief for this booking already exists, so nothing is written twice. 7. Read the saved business fields this automation is locked to: your voice, offer, audience, method, proof, guardrails, and links. 8. If enough context to sound like you, then go to "Research the booker", otherwise go to "Note the gaps". 9. Note which saved fields are empty or thin, so the brief opens with an honest line instead of pretending. 10. Look up the booker from the booking details (their company's public site, from their email domain). 11. If research found anything, then go to "Write the brief with Claude", otherwise go to "Say the research is thin". 12. Note that nothing public was found, so the brief says the research is thin instead of padding. 13. Write the brief and talking points in your saved voice, marking every claim as found (with the source), inferred, or assumed, and never using the words your guardrails forbid. 14. Add one row to the Call briefs list, keyed by the booking id, so a replayed booking never writes a second brief. 15. Post the brief to your team chat before the call. 16. This was a cancellation or reschedule, so no brief is written. The tools: - Cal.com (for "Where do your bookings come from?"). - Anthropic (for "What should research and write the brief?"). I will need you to create the key ANTHROPIC_API_KEY (Anthropic API key). I add the values in Autoploy myself; you only need their names. - Slack (for "Where should the brief land before the call?"). I will need you to create the key SLACK_BOT_TOKEN (Slack bot token), provide the setting SLACK_CHANNEL (Slack channel). I add the values in Autoploy myself; you only need their names. Shared memory: This automation remembers its work in a list called "Call briefs" (Every pre-call brief The Closer has written, one row per booked call.) that I can open in the Context section of my dashboard and see every row. Its columns are: who ("Who"), company ("Company"), call_time ("Call time"), brief ("Brief"), status ("Status"), booking ("Booking"). Before anything runs, make sure the list exists: check with list_collections over my Autoploy MCP, and create it with create_collection (exactly that name, that purpose, and those columns) only if it is missing. Never create a second list under a similar name; reuse this one. The platform never creates a list silently at run time, so a missing list stops the work with a plain refusal. In the automation's code, write rows through ctx.collections.insert with an idempotency key built from the id of the thing the row is about (for example the booking id), so a replayed trigger can never write the same row twice. Read prior rows with ctx.collections.query when the automation must know what it already did. The platform: Connect to my Autoploy MCP server, call get_authoring_guide to read the authoring guide, and build the automation with the Autoploy SDK contract. Model every yes/no fork as a condition step with truthful labeled branches, keep every external side effect idempotent and keyed, and deploy it with deploy_automation. Before you build: Repeat the plan back to me in plain English and ask follow-up questions about anything unclear or ambiguous. Do not guess. Ask me for each key by name when it is time, and never ask me to paste a key anywhere except my own secure setup.