AI Agent Email Blog
Learn about AI agent email — tutorials, use cases, and updates from the LobsterMail team.
Browse by topic
email deliverability for ai agents: what actually matters in 2026
AI agents need dedicated domains, proper authentication, and smart warm-up to land in the inbox. Here's what email deliverability looks like when your sender isn't human.
gmail api 429 too many requests: what it means and how to fix it
A Gmail API 429 error means you've hit Google's rate limits. Here's what triggers it, how to fix it, and when to stop fighting quotas entirely.
langchain gmail tool integration: the full setup guide (and where it breaks)
How to set up the LangChain GmailToolkit, what tools it includes, OAuth gotchas for long-running agents, and when you might want something simpler.
llamaindex reactagent deprecated: how to migrate to the new workflow-based agent
LlamaIndex deprecated ReActAgent, AgentRunner, and FunctionCallingAgent. Here's the full migration path with before/after code examples.
NemoClaw vs LobsterMail: What NVIDIA's Agent Platform Means for Email
How NemoClaw-powered AI agents handle email — and why LobsterMail's programmatic inbox provisioning solves the scaling problems that IMAP-based setups create.
ai agent email bounce handling automation: the complete decision tree
How autonomous AI agents process bounces, suppress bad addresses, and protect domain reputation without human intervention.
AutoGen Email Tool Integration: Every Method Compared
A practical comparison of every way to wire email into AutoGen agents — direct SMTP, Gmail API, MCP endpoints, and LobsterMail — including reply threading and inbound routing.
how to publish a skill on clawhub
A step-by-step guide to publishing your first OpenClaw skill on ClawHub, from account setup to CLI commands to post-launch maintenance.
how to use llama_index agent email tools (and when they fall short)
A practical guide to adding email capabilities to LlamaIndex agents using GmailToolSpec, custom FunctionTools, and dedicated email infrastructure for production.
llm agent prompt injection design patterns: 6 defenses that actually work
Six design patterns protect LLM agents from prompt injection attacks. Here's how each one works, when to use it, and what it costs in practice.
mandrill alternative for ai agents: what actually works
Mandrill wasn't built for AI agents. Here are the alternatives that handle autonomous sending, inbound parsing, and agent-first provisioning in 2026.
agentmail vs resend: which email API actually works for AI agents?
A neutral comparison of AgentMail and Resend for AI agent email. Covers pricing, inboxes, threading, and when each tool is the right choice.