Source: https://cli.nylas.com/guides/manus-email-drafter Guide # Draft Personalized Emails with Manus AI Use Manus AI to draft personalized outbound emails at scale. Provide context and a contact list, review AI-generated drafts, and send with the Nylas CLI across all major email providers. This workflow covers sales outreach, follow-ups, and newsletters. Written by [Aaron de Mello](https://cli.nylas.com/authors/aaron-de-mello) • Senior Engineering Manager Reviewed by [Hazik](https://cli.nylas.com/authors/hazik) Updated April 11, 2026 Verified — CLI 3.1.1 · Gmail, Outlook · last tested April 11, 2026 ## Why use Manus for email drafting Writing personalized emails is time-consuming. A generic template feels impersonal, but hand-crafting every message does not scale. Manus bridges this gap. You give the agent context — previous conversations, company info, a template — and it drafts emails tailored to each recipient. You review every draft before anything gets sent. The Nylas CLI provides the execution layer. Manus uses `nylas email smart-compose` to generate drafts and `nylas email send` to deliver them. The agent handles the iteration; you handle the judgment. ## Prerequisites 1. **A Manus account** with the Nylas CLI Skill installed. Follow the [Manus AI Skills setup guide](https://cli.nylas.com/guides/manus-ai-skills) if you have not done this yet. 2. **Nylas CLI authenticated** with at least one connected grant. Run `nylas auth whoami` to verify. ## Draft a single email The simplest use case is asking Manus to draft one email from a prompt. In the Manus chat, type something like: *"Draft a follow-up email to Sarah about the budget meeting."* Manus translates this into a smart-compose command: ```bash nylas email smart-compose --prompt "Polite follow-up about the Q4 budget discussion. Ask for the updated spreadsheet." ``` The CLI returns a draft. Manus shows it to you for review. No email is sent at this stage. ## Draft from context For better personalization, let Manus read existing threads before composing. Ask: *"Read my last email from Bob and draft a reply accepting his proposal."* Manus first retrieves the message to understand the conversation context: ```bash nylas email read MESSAGE_ID --json ``` Then it composes a contextual reply using the content of Bob's email as input: ```bash nylas email smart-compose --prompt "Reply to Bob accepting his proposal. Reference the timeline he mentioned and confirm the March 20 start date." ``` This two-step pattern — read then compose — produces drafts that feel like genuine replies rather than generic messages. ## Batch drafting For outbound campaigns, give Manus a list of contacts with context. In the Manus chat: *"For each of these 5 contacts, draft a personalized intro email mentioning their company and role: Sarah Chen (VP Engineering, Acme Corp), James Park (CTO, Bolt Labs),..."* Manus iterates through the list, calling smart-compose for each contact with a tailored prompt. Each draft is shown to you individually for approval before moving to the next. For larger lists, break them into batches of 10-20 contacts to keep review manageable and stay within credit limits. ## Review and send After you approve a draft, Manus sends it using the CLI: ```bash nylas email send --to "sarah.chen@acme.com" --subject "Quick intro — Nylas + Acme" --body "Hi Sarah, ..." --yes ``` The `--yes` flag skips the CLI's interactive confirmation prompt (which would hang in the Manus sandbox). Human review happens at the Manus level — you see the draft and approve it before Manus runs the send command. ## Templates You can give Manus a template with placeholders and it fills them in per contact. Paste the template directly in the chat or reference a file: ```text Subject: {{FIRST_NAME}}, quick question about {{COMPANY}} Hi {{FIRST_NAME}}, I noticed {{COMPANY}} recently {{RECENT_EVENT}}. We help teams like yours {{VALUE_PROP}}. Would you be open to a 15-minute call this week? Best, [Your name] ``` Manus replaces each placeholder based on the contact details you provide and adjusts the tone and phrasing so the emails do not read like mail-merge output. ## Safety Always review before sending. The [SKILL.md](https://cli.nylas.com/guides/manus-ai-skills) should include an explicit instruction that Manus must never send without confirmation. This is the most important safeguard in the workflow. - **Human-in-the-loop**: Manus shows each draft and waits for your "yes" before sending. If you say "no" or ask for changes, it revises the draft. - **The `--yes` flag**: This bypasses the CLI's built-in confirmation, not yours. It exists because interactive prompts hang in the Manus sandbox. Your approval still gates every send. - **Batch safety**: For batch sends, review at least the first 2-3 drafts carefully. If the quality and tone are consistent, you can approve the rest more quickly. ## FAQ ### How many emails can Manus draft in one batch? There is no hard limit on drafts per batch. Manus processes each contact sequentially using smart-compose. Practical limits depend on your Manus credit balance and Nylas API rate limits. For large lists (50+), break them into smaller batches of 10-20 contacts. ### How good is the personalization quality? It depends on the context you provide. Giving Manus access to previous email threads, company details, or a detailed template produces significantly better results than a generic prompt. The smart-compose command uses the Nylas AI layer, which is tuned for professional email tone. ### Will Manus send emails without my approval? Not if your SKILL.md includes the instruction to always confirm before sending. Manus follows the rules in the Skill file. The review step is built into the workflow — Manus shows each draft and waits for your approval before running the send command. ## Next steps - [Manus AI Inbox Zero](https://cli.nylas.com/guides/manus-inbox-zero): triage incoming email before drafting replies - [Sync Email to CRM with Manus AI](https://cli.nylas.com/guides/manus-ai-crm-email): extract contacts from your inbox for CRM import - [Create a Manus Skill for Email and Calendar](https://cli.nylas.com/guides/manus-ai-skills): set up the Nylas CLI Skill from scratch - [Command reference](https://cli.nylas.com/docs/commands): every CLI flag, subcommand, and example