# MCPmailer blog Writing about email for AI agents: identity, deliverability, oversight, and MCP. ## Writing documentation an agent can actually answer from 6 August 2026 · 4 min read Your agent answers from your help centre, which was written for people who are already reading it. What to change so retrieval finds the right passage and the answer is groundable. Read: https://mcpmailer.com/blog/writing-docs-agents-can-use.md ## A test suite for an agent that writes email 6 August 2026 · 3 min read How to write the golden thread harness in code: fixtures, assertions on actions rather than wording, a fake mail client, and the adversarial cases worth pinning. Read: https://mcpmailer.com/blog/testing-email-agents-with-vitest.md ## Rotating an agent's key without breaking it 6 August 2026 · 4 min read Overlap windows, where keys actually leak from, what to do when one is exposed, and why the agent should never be the thing that holds the only copy. Read: https://mcpmailer.com/blog/rotating-agent-credentials.md ## One brand, forty branches, and an inbox that has to know which 6 August 2026 · 4 min read Customers write to head office about a specific branch. How to route by location, whether each site gets its own agent, and who owns the reply when the answer is local. Read: https://mcpmailer.com/blog/multi-location-email-routing.md ## The week between Christmas and New Year 6 August 2026 · 4 min read Coverage gaps are predictable, and so is what arrives during them. What an agent should do when nobody is available, what it must not promise, and how to hand back a backlog. Read: https://mcpmailer.com/blog/holiday-coverage-email-agents.md ## The forwarded message problem 6 August 2026 · 4 min read A forward carries a stranger's words, a broken reply-to, and instructions the agent should not follow. How to parse one, who to answer, and why forwards break authentication. Read: https://mcpmailer.com/blog/handling-forwarded-email.md ## The vocabulary of agent email, defined 6 August 2026 · 5 min read SPF, DKIM, DMARC, MCP, threading, cold sends, suppression, idempotency. Every term that appears when you put an agent on an inbox, with what it means and why it matters. Read: https://mcpmailer.com/blog/email-agent-glossary.md ## Getting DNS records published when you do not control DNS 6 August 2026 · 4 min read The technical setup takes ten minutes and the organisational one takes three weeks. Delegation, CNAME flattening, apex limitations, and how to hand records to whoever owns the zone. Read: https://mcpmailer.com/blog/dns-delegation-for-agent-domains.md ## It sent the wrong thing. Now what 6 August 2026 · 4 min read A wrong reply has five common causes and each leaves a different trace. How to find which one it was, in order, without reading message bodies out of a log. Read: https://mcpmailer.com/blog/debugging-an-email-agent.md ## The HR inbox, where the same policy is explained daily 6 August 2026 · 4 min read Leave, payroll, benefits, and policy questions repeat endlessly and arrive alongside things that must never be automated. Where the line sits, and why confidentiality changes the design. Read: https://mcpmailer.com/blog/ai-agent-hr-employee-questions.md ## Where is my shipment, asked forty times a day 6 August 2026 · 4 min read Freight runs on status chasing across carriers, ports, and time zones. What an agent answers from tracking data, what it must never commit to, and why exceptions are the real work. Read: https://mcpmailer.com/blog/ai-agent-freight-logistics.md ## Practice email, where the reception phone is already ringing 6 August 2026 · 4 min read Dental, veterinary, and therapy practices run on appointments and reminders. What an agent handles for a small reception team, and the clinical line it must never approach. Read: https://mcpmailer.com/blog/ai-agent-clinic-appointments.md ## Working out who actually sent that message 5 August 2026 · 4 min read The From header is a claim. What authentication results actually prove, how to check a sender against what you already know, and why identity should never authorise an action. Read: https://mcpmailer.com/blog/verifying-who-emails-your-agent.md ## You already send with a transactional provider. Now it has to receive 5 August 2026 · 4 min read Resend, Postmark, SendGrid, and Mailgun send well. The moment an agent has to read replies you need threading, contact memory, and an inbox, which is a different product. Read: https://mcpmailer.com/blog/transactional-provider-to-agent-inbox.md ## Getting a damaged sending domain back 5 August 2026 · 4 min read Reputation falls fast and returns slowly. How to tell what actually broke, whether to repair or start a new subdomain, and the schedule that gets you back without a second collapse. Read: https://mcpmailer.com/blog/sending-reputation-recovery.md ## Putting an agent on an inbox that already has traffic 5 August 2026 · 5 min read Starting an agent on a new address is easy. Moving one that customers already write to, with expectations and history attached, needs a different plan. Read: https://mcpmailer.com/blog/migrating-live-inbox-to-agent.md ## Making an agent answer only what it can prove 5 August 2026 · 5 min read The expensive failure is not a wrong tone, it is a fluent invented fact a customer believes. How to require a source for every claim, and what to do when there is not one. Read: https://mcpmailer.com/blog/grounding-and-citation-email-agents.md ## Building the agent's mail layer yourself on SES 5 August 2026 · 4 min read SES is cheap and capable, and an agent needs a dozen things it does not provide. An honest list of what you would build, how long it takes, and when doing it is the right call. Read: https://mcpmailer.com/blog/building-email-agent-on-ses-directly.md ## Renewals, lapses, and the members who meant to come back 5 August 2026 · 4 min read Most lapsed members did not decide to leave, they missed a card expiry and a reminder. What an agent should send, when to stop, and why cancellation is the conversation worth handling well. Read: https://mcpmailer.com/blog/ai-agent-membership-renewals.md ## Taking the details of a claim without deciding anything 5 August 2026 · 4 min read First notification of loss is structured data collection arriving as prose, at the worst moment in a customer's year. What an agent gathers, and why coverage decisions must stay untouched. Read: https://mcpmailer.com/blog/ai-agent-insurance-claims-intake.md ## Booking enquiries that arrive while the desk is busy 5 August 2026 · 4 min read Availability, rates, and the twenty questions before a booking arrive at the worst moments. What an agent answers from the property management system, and what stays with the desk. Read: https://mcpmailer.com/blog/ai-agent-hospitality-bookings.md ## Admissions email, where every applicant thinks they are the only one 5 August 2026 · 4 min read Admissions inboxes spike at deadlines and every message is important to the person who sent it. What an agent answers, what it must never decide, and the safeguarding line. Read: https://mcpmailer.com/blog/ai-agent-education-admissions.md ## Quote requests that arrive while you are on a roof 5 August 2026 · 4 min read Trades lose work by not replying, not by quoting badly. What an agent can do with an enquiry before anyone reaches a laptop, and why it must never put a number on anything. Read: https://mcpmailer.com/blog/ai-agent-construction-quoting.md ## Where an agent fits if you already run a helpdesk 5 August 2026 · 4 min read You already have a ticketing system and it is not going anywhere. Four ways to add an agent around it, which one to pick, and how to avoid two systems answering the same customer. Read: https://mcpmailer.com/blog/agent-alongside-helpdesk.md ## When an email agent is the wrong answer 4 August 2026 · 5 min read Agents are a poor fit for several common situations, and finding out after launch is expensive. The cases where the answer is a form, a phone call, better documentation, or nothing. Read: https://mcpmailer.com/blog/when-not-to-use-an-email-agent.md ## The weekly half hour that keeps an agent honest 4 August 2026 · 4 min read Dashboards tell you an agent is running. Reading fifteen threads tells you whether it is any good. What to sample, what to look for, and what to do with what you find. Read: https://mcpmailer.com/blog/weekly-review-email-agents.md ## Email tools for an agent built with the Vercel AI SDK 4 August 2026 · 3 min read Define email as typed tools, keep the loop bounded, and put the untrusted message where it cannot reach a privileged call. A working shape with the parts that matter explained. Read: https://mcpmailer.com/blog/vercel-ai-sdk-email-agent.md ## No SDK for your language, and it does not matter 4 August 2026 · 3 min read The whole surface is an HTTP API with a bearer token. What the four calls look like in Go, Ruby, and PHP, and the three things to get right whatever you write it in. Read: https://mcpmailer.com/blog/email-agent-any-language.md ## Designing the tools your email agent calls 4 August 2026 · 4 min read Most agent failures trace to tool design rather than the model: too many tools, vague names, errors that invite retries, and results that hand the model raw text. What good looks like. Read: https://mcpmailer.com/blog/designing-tools-for-email-agents.md ## Sending the model less than you think it needs 4 August 2026 · 4 min read Every message you pass to a model is data leaving your systems. What to strip, what to summarise, what never to send at all, and why the cheapest version is usually the safest. Read: https://mcpmailer.com/blog/data-minimisation-email-agents.md ## The requirements Gmail and Yahoo now enforce 4 August 2026 · 4 min read Authentication, one-click unsubscribe, and a complaint rate ceiling are now enforced rather than encouraged. What each one means, and why agent mail usually passes without effort. Read: https://mcpmailer.com/blog/bulk-sender-requirements.md ## A recruitment agency runs on email, and most of it is logistics 4 August 2026 · 4 min read Agency recruiters work two inboxes at once. What an agent can take on each side, why candidate and client agents should be separate, and the lines it must never cross. Read: https://mcpmailer.com/blog/ai-agent-recruitment-agency-email.md ## Client intake for a firm that bills by the hour 4 August 2026 · 4 min read Law and accounting firms lose billable time to intake, document chasing, and status questions. What an agent can take, and the bright line around advice it must never cross. Read: https://mcpmailer.com/blog/ai-agent-professional-services-intake.md ## Thanking donors properly, at the volume you actually have 4 August 2026 · 4 min read Small teams cannot personally answer every donor, so most send a receipt and nothing else. What an agent can take on, and why the thank-you is the one thing worth keeping human-shaped. Read: https://mcpmailer.com/blog/ai-agent-nonprofit-donor-email.md ## Support when both people emailing you are your customers 4 August 2026 · 4 min read Marketplaces answer two sides of every dispute, and each side thinks you work for them. How an agent handles order questions, when it must never adjudicate, and why one thread per side matters. Read: https://mcpmailer.com/blog/ai-agent-marketplace-support.md ## The internal helpdesk nobody wanted to staff 4 August 2026 · 4 min read Internal support is the easiest agent to justify and the easiest to get wrong, because the requests that look routine are the ones that grant access. What to automate and what to refuse. Read: https://mcpmailer.com/blog/ai-agent-internal-it-helpdesk.md ## Transactional, marketing, conversational: which one is your agent sending 3 August 2026 · 4 min read The three categories have different infrastructure, different rules, and different consequences for mixing them. Where agent mail sits, and why sending it through a campaign tool goes wrong. Read: https://mcpmailer.com/blog/transactional-vs-conversational-email.md ## Every tool you connect is a tool your inbox can be talked into using 3 August 2026 · 4 min read An agent with an inbox and a dozen connected tools has a supply chain. What a hostile or compromised server can do, why email raises the stakes, and how to connect one safely. Read: https://mcpmailer.com/blog/third-party-mcp-server-risk.md ## Subject lines an agent should write, and the ones it should leave alone 3 August 2026 · 4 min read The subject is the one line that decides whether anything else gets read, and the one field agents most often break. When to keep it, when to write it, and what makes a good one. Read: https://mcpmailer.com/blog/subject-lines-for-agent-email.md ## What your team does once the agent handles the easy half 3 August 2026 · 5 min read An agent changes the shape of the work rather than the amount, and the leftover pile is harder than the average was. How roles, staffing levels, and hiring change, honestly. Read: https://mcpmailer.com/blog/staffing-around-email-agents.md ## Setting response time targets you can actually keep 3 August 2026 · 4 min read An agent answers in seconds, which makes it tempting to promise something you cannot hold when it escalates. How to set targets per category, measure the tail, and design the fallback. Read: https://mcpmailer.com/blog/response-time-targets-agent-email.md ## Connecting a hosted client that cannot hold a key 3 August 2026 · 4 min read Some clients cannot store an API key, so they authorize as you instead. How the OAuth flow works, what the scopes cover, why a token needs an agent header, and when to prefer a key. Read: https://mcpmailer.com/blog/oauth-for-hosted-agent-clients.md ## Running an email agent on ordinary Node infrastructure 3 August 2026 · 4 min read Not everything runs on the edge. The same agent as a Node service: webhook endpoint, a queue, per-thread locking in Postgres, and the deployment details that decide whether it survives a burst. Read: https://mcpmailer.com/blog/email-agent-on-node.md ## When an agent should send one email instead of nine 3 August 2026 · 4 min read Agents can send a message per event, and usually should not. How to decide between immediate, batched, and digest delivery, and how to implement the batching without losing urgency. Read: https://mcpmailer.com/blog/digest-vs-individual-emails.md ## Answering property inquiries before the lead goes cold 3 August 2026 · 4 min read Portal inquiries arrive in bursts and decay in hours. How an agent answers the same six questions, books viewings, and hands over the moment a conversation becomes a negotiation. Read: https://mcpmailer.com/blog/ai-agent-real-estate-inquiries.md ## Running an event's inbox without a spreadsheet 3 August 2026 · 4 min read Registrations, dietary requirements, transfers, cancellations, and the flood of questions in the last 48 hours. What an agent handles, what it must not decide, and how to survive the week of the event. Read: https://mcpmailer.com/blog/ai-agent-event-registration-email.md ## Keeping clients informed without a weekly status meeting 3 August 2026 · 4 min read Agencies lose hours to status emails, chasing approvals, and collecting assets. An agent can run all three, provided it never negotiates scope or speaks for the team on judgement calls. Read: https://mcpmailer.com/blog/ai-agent-agency-client-updates.md ## What goes at the bottom of an agent's email 3 August 2026 · 4 min read The signature is where disclosure, the route to a human, and your legal obligations live. What belongs there, what does not, and why an agent's footer differs from a person's. Read: https://mcpmailer.com/blog/agent-email-signatures-and-disclosure.md ## Where each piece of an agent's state belongs 2 August 2026 · 5 min read Thread, contact, note, vault, your database, and the agent's own storage each hold a different kind of state. Putting something in the wrong one is why agents drift, repeat themselves, and contradict your systems. Read: https://mcpmailer.com/blog/where-agent-email-state-lives.md ## What DMARC reports are actually telling you 2 August 2026 · 5 min read Aggregate reports are ugly XML that answer one useful question: who else sends as your domain. How to read them, what a failure means, and when it is safe to move to reject. Read: https://mcpmailer.com/blog/reading-dmarc-reports.md ## Giving every customer their own agent inbox 2 August 2026 · 5 min read If your product runs an agent for each customer, each one needs its own address. How to provision at signup, keep tenants isolated, handle domains, and keep one tenant's mistake from reaching the others. Read: https://mcpmailer.com/blog/inbox-per-tenant-architecture.md ## Knowing where your agent's mail actually lands 2 August 2026 · 5 min read Delivered is not the same as seen. Postmaster tools, seed accounts, and the handful of numbers that tell you placement is slipping before a customer does. Read: https://mcpmailer.com/blog/deliverability-monitoring-seed-testing.md ## When someone sends email pretending to be you 2 August 2026 · 5 min read DMARC enforcement stops spoofing of your exact domain and does nothing about lookalikes. What each defence actually covers, what to monitor, and what to do the day it happens. Read: https://mcpmailer.com/blog/brand-impersonation-lookalike-domains.md ## Deciding what each agent is allowed to see 2 August 2026 · 5 min read Least privilege inside a workspace: what an identity can reach by default, how contact, note, and vault grants work, which human roles can change them, and how to design the smallest useful agent. Read: https://mcpmailer.com/blog/agent-permissions-and-grants.md ## Chasing quotes and vendors without becoming the bottleneck 1 August 2026 · 5 min read Procurement email is coordination, not judgement: requesting quotes, chasing responses, normalising what comes back, and keeping the decisions with a person. Read: https://mcpmailer.com/blog/vendor-procurement-email-agent.md ## An email agent without writing a service 1 August 2026 · 4 min read Wire an inbox to an automation tool: receive with a webhook, decide with an LLM step, reply through one HTTP call, and the four things to get right so it does not double-send. Read: https://mcpmailer.com/blog/no-code-email-agent-automation.md ## Answering email in the language it arrived in 1 August 2026 · 5 min read Language is the cheapest coverage win an email agent offers, and the easiest place to embarrass yourself. Detection, register, what to never translate, and when to escalate to a person who speaks it. Read: https://mcpmailer.com/blog/multilingual-email-agent.md ## Adding email to a LangChain or LangGraph agent 1 August 2026 · 4 min read Wire an inbox into a LangChain or LangGraph agent: MCP where the adapter exists, plain tools where it does not, and where the email work belongs in a graph. Read: https://mcpmailer.com/blog/langchain-email-agent.md ## Answering "where is my order" without a human 1 August 2026 · 5 min read WISMO is the highest volume, lowest variety mail in retail. How an agent answers from carrier and order data, handles the exceptions honestly, and knows which cases are refunds in disguise. Read: https://mcpmailer.com/blog/ecommerce-order-status-agent.md ## When an agent and a person are on the same thread 1 August 2026 · 5 min read Cc, reply-all, and handover are where agents embarrass you in front of colleagues and customers. The rules that keep a shared thread readable, and who should be speaking when. Read: https://mcpmailer.com/blog/agents-and-humans-on-one-thread.md ## Dedicated sending IPs, and why most agents should not want one 31 July 2026 · 4 min read A dedicated IP sounds like an upgrade and is a liability below a certain volume. What IP reputation actually does now, when a dedicated one helps, and what to fix instead. Read: https://mcpmailer.com/blog/dedicated-vs-shared-ip-agent-email.md ## Onboarding email that notices when someone is stuck 31 July 2026 · 4 min read Onboarding sequences send on a schedule and ignore the answer. An agent reads the reply, checks what the customer has actually done, and stops when they are up and running. Read: https://mcpmailer.com/blog/ai-agent-customer-onboarding-email.md ## Keeping an agent inbox usable after ten thousand messages 31 July 2026 · 4 min read An inbox nobody prunes becomes a list nobody can use. How agents should search, what archiving is actually for, and why unread is a signal rather than a status. Read: https://mcpmailer.com/blog/agent-inbox-search-and-archive.md ## When the agent sends something it should not have 31 July 2026 · 5 min read A runbook for the hour after an agent sends the wrong thing: stopping it, working out the blast radius, telling the recipients, and the postmortem that stops it repeating. Read: https://mcpmailer.com/blog/agent-email-incident-response.md ## Where the mail actually goes: data flow for an agent inbox 31 July 2026 · 5 min read The question procurement asks and most teams cannot answer: what leaves your infrastructure when an agent reads and sends mail, who processes it, and which parts you control. Read: https://mcpmailer.com/blog/agent-email-data-residency.md ## What happens when forty messages arrive at once 31 July 2026 · 5 min read Agent email is bursty. How to handle a queue of inbound mail without double-replying, racing on one thread, or spending your daily allowance in ninety seconds. Read: https://mcpmailer.com/blog/agent-email-concurrency-bursts.md ## An email agent in TypeScript, from empty folder to first reply 30 July 2026 · 3 min read A working TypeScript agent with its own inbox: install, send, read the thread, wait for a reply, handle refusals, and receive inbound mail without a polling loop. Read: https://mcpmailer.com/blog/typescript-email-agent-sdk.md ## Giving an OpenAI-based agent an email address 30 July 2026 · 4 min read Connect a GPT-based agent to a real inbox: MCP where the tooling supports it, plain function tools where it does not, and the guardrails to set before it emails anyone. Read: https://mcpmailer.com/blog/openai-agent-email-tutorial.md ## Deciding what reaches your agent in the first place 30 July 2026 · 5 min read Whitelist or blacklist, exact-address rules that beat domain rules, and how to route one category of real mail to an agent without redirecting your whole inbox. Read: https://mcpmailer.com/blog/inbound-email-routing-agents.md ## The first week with an email agent, day by day 30 July 2026 · 5 min read A day-by-day plan for the week that decides whether an email agent works: what to set up, what to read, what to fix, and what to hand it next. Read: https://mcpmailer.com/blog/first-week-with-an-email-agent.md ## Appointment reminders that can handle the reply 30 July 2026 · 4 min read A reminder that cannot process "can we move it to Thursday" is half a feature. How an agent confirms, reminds, reschedules, and knows when to stop chasing. Read: https://mcpmailer.com/blog/ai-agent-appointment-reminders.md ## Driving an agent's inbox from the terminal 30 July 2026 · 4 min read The CLI is the fastest way to see what an agent is doing: list its mail, read a thread, send a test message, and script the checks worth running in CI. Read: https://mcpmailer.com/blog/agent-inbox-from-the-terminal.md ## The security review questions an email agent should survive 29 July 2026 · 5 min read The questions a security team will ask before an agent touches customer mail, why each one is asked, and what a good answer looks like when the agent is the thing under review. Read: https://mcpmailer.com/blog/security-review-email-agent.md ## A personal AI email assistant that is not just a draft button 29 July 2026 · 5 min read Most AI email assistants sit inside your mailbox and draft. An assistant with its own address can be delegated to instead: cc it, forward to it, and let it handle the thread. Read: https://mcpmailer.com/blog/personal-ai-email-assistant.md ## Do spam filters catch AI-written email? 29 July 2026 · 4 min read Filters do not detect authorship, they detect patterns. What actually triggers them in agent-sent mail, which AI writing habits look like bulk, and how to write so a real message reads as one. Read: https://mcpmailer.com/blog/does-ai-written-email-get-filtered.md ## Bounces, complaints, and the suppression list your agent cannot argue with 29 July 2026 · 5 min read Hard bounces, soft bounces, complaints, and unsubscribes each need a different response. What the platform enforces, what your agent must do with the rest, and why suppression belongs below the model. Read: https://mcpmailer.com/blog/bounces-suppression-agent-email.md ## Making your product readable by the agents that will recommend it 29 July 2026 · 5 min read Agents now evaluate and integrate products by reading documentation. What llms.txt is, why serving markdown beside HTML matters, and how to structure docs an agent can act on. Read: https://mcpmailer.com/blog/agent-readable-website-llms-txt.md ## Keys, scope, and what to do the day one leaks 29 July 2026 · 5 min read An agent key is a credential that can send email as your company. How to scope, store, rotate, and revoke keys, and the hour-one checklist for a key that has leaked. Read: https://mcpmailer.com/blog/agent-api-key-security.md ## What to monitor when an agent is answering your mail 28 July 2026 · 5 min read The signals that catch an email agent going wrong before a customer does: send classification, first contact with new domains, silence, bounce movement, and the traces worth keeping per thread. Read: https://mcpmailer.com/blog/monitoring-email-agents-in-production.md ## Moving an agent off the Gmail API onto its own inbox 28 July 2026 · 5 min read A migration plan for agents built on Gmail or Graph: what breaks at scale, how to move one flow at a time without a cutover, and what to do about history, tokens, and the old address. Read: https://mcpmailer.com/blog/migrate-gmail-api-to-agent-inbox.md ## Designing approval that a human will still read in week three 28 July 2026 · 5 min read Approval queues fail by being too full to read. How to choose what needs a person, design the review itself, set a timeout for unreviewed items, and know when to remove the gate. Read: https://mcpmailer.com/blog/human-in-the-loop-email-approval.md ## How email threading actually works, and why agents break it 28 July 2026 · 5 min read Threading is a convention held together by three headers and a lot of client guesswork. What they do, how agents break conversations, and why a broken thread is worse than a wrong answer. Read: https://mcpmailer.com/blog/email-threading-for-agents.md ## The compliance floor for email an AI agent sends 28 July 2026 · 6 min read An agent sending on your behalf is you sending. What CAN-SPAM, GDPR, PECR, and CASL require, how unsubscribe and suppression have to work, and what to do about disclosure and retention. Read: https://mcpmailer.com/blog/ai-email-compliance-gdpr-canspam.md ## Test inboxes for agent development, without mocking email 27 July 2026 · 5 min read How to test an email agent properly: disposable inboxes per run, replaying real threads, asserting on what was sent, and the difference between a sandbox that traps mail and a real one that delivers. Read: https://mcpmailer.com/blog/test-email-addresses-for-agents.md ## Sending and receiving email from a Python agent 27 July 2026 · 4 min read A working Python agent with its own inbox: install the SDK, send a first message, read threads, wait for replies, handle refusals, and receive inbound mail without a polling loop. Read: https://mcpmailer.com/blog/python-email-agent-tutorial.md ## Attachments: what an agent should do with the file in the email 27 July 2026 · 5 min read Invoices, contracts, and CVs arrive as attachments. How agents receive and send files over email, what to extract before the model reads, and the safety rules that keep a file from becoming an exploit. Read: https://mcpmailer.com/blog/email-attachments-for-agents.md ## What email actually costs when your agents send it 27 July 2026 · 5 min read Per-message pricing, per-inbox pricing, storage, and the hidden costs nobody budgets for: tokens spent parsing mail, engineering time on threading, and the price of a damaged sending domain. Read: https://mcpmailer.com/blog/cost-of-email-for-ai-agents.md ## Build an email agent on the Cloudflare Agents SDK 27 July 2026 · 4 min read A working email agent as a Durable Object: connect the MCP server, react to inbound mail through a webhook, hold a thread across days, and keep the whole thing idempotent. Read: https://mcpmailer.com/blog/build-email-agent-agents-sdk.md ## Login codes, 2FA, and verification email an agent has to handle 27 July 2026 · 5 min read Agents that sign up for services hit email verification and two-factor prompts. How to receive codes in an agent inbox, clear TOTP without a human relaying digits, and not build a phishing target. Read: https://mcpmailer.com/blog/agent-verification-codes-email.md ## An agent that signs itself up for its own inbox 27 July 2026 · 5 min read Autonomous agents need addresses before a human is available to create one. How self-signup works, what a claim URL is for, and the guardrails that make agent-provisioned inboxes safe to allow. Read: https://mcpmailer.com/blog/agent-self-signup-email.md ## Giving an email agent memory of the people it talks to 27 July 2026 · 5 min read The difference between an agent that feels competent and one that feels robotic is memory. How to use contacts, recorded facts, and notes so the second conversation starts where the first ended. Read: https://mcpmailer.com/blog/agent-memory-contacts-email.md ## Webhooks or polling: how your agent should learn that mail arrived 26 July 2026 · 5 min read A practical comparison of webhook-driven and polling agents, with signature verification, retry and idempotency handling, and the hybrid most production deployments end up running. Read: https://mcpmailer.com/blog/webhooks-vs-polling-agent-email.md ## Long-running email conversations: waiting, timeouts, and resuming 26 July 2026 · 5 min read Email conversations take days; agent runs take seconds. The patterns for holding a thread open, choosing timeouts, resuming with context, and knowing when to stop following up. Read: https://mcpmailer.com/blog/wait-for-reply-long-running-agents.md ## Prompt injection by email, and how to build an agent that survives it 26 July 2026 · 8 min read An email inbox is an untrusted input channel anyone can write to. The injection patterns agents actually get hit with, and the architectural defences that work better than telling the model to be careful. Read: https://mcpmailer.com/blog/prompt-injection-email-agents.md ## Running several email agents in one workspace 26 July 2026 · 5 min read How to structure a workspace once you have more than one agent: address naming, what to share and what to isolate, handoffs between agents, and the failure modes that only appear at three or four. Read: https://mcpmailer.com/blog/multi-agent-email-workspace.md ## How to tell whether your email agent is any good 26 July 2026 · 6 min read Deflection rate flatters everyone. The metrics that expose a bad email agent, how to build a golden thread suite, and how to run regressions before a prompt change reaches customers. Read: https://mcpmailer.com/blog/evaluating-ai-email-agents.md ## A warmup plan for a new agent sending domain 26 July 2026 · 6 min read A new sending subdomain has no reputation and no reputation is treated as suspicious. Build a week by week ramp, understand which traffic warms fastest, and know when to hold. Read: https://mcpmailer.com/blog/email-warmup-plan-agents.md ## The system prompt for an email agent, annotated 26 July 2026 · 5 min read A complete, copyable system prompt for an agent with its own inbox, with the reasoning behind every section: identity, thread discipline, escalation, refusal handling, and tone. Read: https://mcpmailer.com/blog/email-agent-system-prompt.md ## SPF, DKIM, and DMARC for an agent subdomain, with a generator 26 July 2026 · 6 min read Generate the DNS records an agent sending subdomain needs, check a real message's Authentication-Results header, and understand alignment, which is the part that actually fails. Read: https://mcpmailer.com/blog/dmarc-record-generator-agents.md ## How to let Claude send and receive email 25 July 2026 · 5 min read Connect Claude to a real inbox in a few minutes with an MCP email server: config, the first send, holding a conversation, and the limits worth knowing before you let a model email people. Read: https://mcpmailer.com/blog/how-to-make-claude-send-email.md ## Automating invoice chasing with an AI agent that reads the replies 25 July 2026 · 6 min read Dunning sequences stop working the moment a customer replies. Here is how an AI agent chases overdue invoices by email, reads the answers, resolves disputes, and escalates the ones that need a person. Read: https://mcpmailer.com/blog/automated-invoice-chasing-ai-agent.md ## An AI agent for lead follow-up email that does not become spam 25 July 2026 · 6 min read How to use an AI agent to answer inbound leads by email in seconds, qualify in thread, book meetings, and hand over to a human, without turning your domain into a cold outreach liability. Read: https://mcpmailer.com/blog/ai-sales-agent-lead-follow-up.md ## Automating interview scheduling email with an AI recruiting agent 25 July 2026 · 6 min read How to hand interview coordination to an AI agent over email: proposing times across panels, handling reschedules, chasing silence, and keeping candidates informed without a scheduling link. Read: https://mcpmailer.com/blog/ai-interview-scheduling-email-agent.md ## How to build an AI agent that answers customer support emails 25 July 2026 · 7 min read A step by step guide to putting an AI agent on your support inbox: routing, thread context, escalation rules, refund limits, and what to measure in the first month. Read: https://mcpmailer.com/blog/ai-agent-customer-support-email.md ## Agent inbox or shared mailbox: which one should your agent use 24 July 2026 · 6 min read Four ways to give an AI agent email access: shared IMAP credentials, a forwarding alias, the Gmail or Graph API, and a dedicated agent identity. What each costs in scope, threading, and oversight. Read: https://mcpmailer.com/blog/agent-inbox-vs-shared-mailbox.md ## Keeping an autonomous agent from becoming a spam problem 22 July 2026 · 6 min read Autonomy plus email is a reputation and legal risk. The controls that actually work: send classification and tripwires, allowances, narrow approval, per-agent scope, mail rules, and an audit trail someone reads. Read: https://mcpmailer.com/blog/stopping-agent-email-spam.md ## Deliverability for agent senders: SPF, DKIM, DMARC, and behaviour 20 July 2026 · 7 min read Why agent-sent email lands in spam, how to authenticate a sending subdomain with SPF, DKIM, and DMARC, how to warm it, and which sending behaviours mailbox providers actually punish. Read: https://mcpmailer.com/blog/agent-email-deliverability.md ## What an MCP email server is, and how to connect one 15 July 2026 · 6 min read An MCP email server exposes send, read, thread, and wait tools over Model Context Protocol. What the tool surface looks like, how scoping and rate limits work, and how to connect Claude, GPT, and other clients. Read: https://mcpmailer.com/blog/mcp-email-server.md ## How to give an AI agent its own email address 8 July 2026 · 8 min read What it takes to put an AI agent on a real email address: identity, threading, waiting for replies, memory between runs, deliverability, and the oversight that keeps it from becoming a spam problem. Read: https://mcpmailer.com/blog/email-for-ai-agents.md Docs: https://mcpmailer.com/docs.md