Ready-to-run AI agent recipes

1. Trigger: new email in Gmail
2. Send subject+body to OpenAI step with prompt below
3. Apply returned label; save draft for 'Reply'
4. Daily 8am digest of Urgent to Slack

Classify this email as URGENT, REPLY, FYI or JUNK. Return JSON {label, reason, draft_reply}. Draft only if REPLY: under 120 words, my tone is direct and friendly. Email: {{body}}

1. Add company/URL to a Sheet row
2. Agent researches site, news, LinkedIn
3. Writes brief + personalized opener back to row

Research {{company}}. Return: what they sell, ICP, recent news (with links), 2 likely pains our product [X] solves, and a 2-sentence cold opener referencing something specific.

1. Trigger: transcript ready
2. LLM extracts decisions & tasks
3. Create tasks in Asana/Notion with due dates
4. Post summary to Slack channel

From this transcript extract: decisions, action items (owner, task, due date if stated), open questions. Return JSON. Transcript: {{transcript}}

1. Trigger: new video/RSS item
2. Fetch transcript
3. Claude writes 3 formats
4. Save to Notion for approval

Turn this into: an X thread (6 posts, strong hook), a LinkedIn post (180 words), and a 60-word newsletter blurb. Keep my voice: [examples]. Source: {{text}}

1. Upload docs/FAQ/help center
2. Set escalation rules
3. Embed widget; review weekly unanswered questions

You are support for [Company]. Answer only from the provided docs. If not covered or customer is upset, say you'll connect them to a human and collect their email.

1. Trigger: email with PDF attachment
2. Extract vendor, amount, date, VAT
3. Append to Sheet / create bill in QuickBooks
4. Flag duplicates

Extract from this invoice text: vendor, invoice_number, date, currency, net, VAT, total. Return JSON; null if missing. Text: {{pdf_text}}

1. Monitor pricing & changelog pages
2. On change, send diff to LLM
3. Weekly summary email

Summarize what changed on {{competitor}}'s page and why it may matter to us (pricing, positioning, features). 3 bullets max. Diff: {{diff}}

1. Pull new comments
2. Classify: question / praise / complaint / spam
3. Draft reply; human approves

Draft a reply to this comment in our brand voice ([voice]). Questions: answer briefly. Complaints: empathize, move to DM. Spam: return SKIP. Comment: {{comment}}