Blog

Page 3 of 10.

An inbox where most of the mail is not a fit for automation

4 August 2026 · 5 min read

When an email agent is the wrong answer

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.

Fifteen threads pulled from a week and read end to end

4 August 2026 · 4 min read

The weekly half hour that keeps an agent honest

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.

Typed tools bound to a model call, with the inbox behind them

4 August 2026 · 3 min read

Email tools for an agent built with the Vercel AI SDK

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.

One HTTP API reached from several languages

4 August 2026 · 3 min read

No SDK for your language, and it does not matter

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.

A small set of tools around one agent

4 August 2026 · 4 min read

Designing the tools your email agent calls

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.

A thread trimmed to what the answer actually needs before it leaves

4 August 2026 · 4 min read

Sending the model less than you think it needs

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.

Three requirements checked before mail is accepted

4 August 2026 · 4 min read

The requirements Gmail and Yahoo now enforce

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.

Two conversations, candidate and client, handled from separate identities

4 August 2026 · 4 min read

A recruitment agency runs on email, and most of it is logistics

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.

Intake, document chasing, and status handled around the work that must stay human

4 August 2026 · 4 min read

Client intake for a firm that bills by the hour

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.

A donation acknowledged with something specific rather than a receipt

4 August 2026 · 4 min read

Thanking donors properly, at the volume you actually have

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.

Two sides of one order, each in their own thread

4 August 2026 · 4 min read

Support when both people emailing you are your customers

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.

An internal request answered from documentation, with an access request routed away

4 August 2026 · 4 min read

The internal helpdesk nobody wanted to staff

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.