using-agentmail

using-agentmail is a skill for Claude Code, Codex from jtsternberg/claude-plugins. It costs 92 tokens per session (6,566 once invoked), scanned A, original, MIT.

An email service for AI agents, provided through the AgentMail API and command-line tool. It gives an agent its own inbox for sending, receiving, replying to, forwarding, and drafting email.

In plain words
What is it for?
Create agent inboxes, send and read messages, manage threads and drafts, and exchange work with people or other agents by email.
Why use it?
It removes the need to handle agent email through a person’s mailbox or raw mail-server protocols. It keeps the agent’s mail identity separate from the user’s Gmail.

Skill for Claude CodeCodex

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the agentmail plugin — 5 skills, 2 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/jtsternberg/claude-plugins/using-agentmail
Any agent
npx skills add jtsternberg/claude-plugins --skill using-agentmail
Clone the repo
git clone --depth 1 https://github.com/jtsternberg/claude-plugins

Made for: Claude Code, Codex.

Or install agentmail, the plugin that ships this one along with the rest of its 5 skills, 2 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for using-agentmail

README.md
[![agentmods](https://agentmods.dev/badge/skills/jtsternberg/claude-plugins/using-agentmail.svg)](https://agentmods.dev/skills/jtsternberg/claude-plugins/using-agentmail)
Your own site
<a href="https://agentmods.dev/skills/jtsternberg/claude-plugins/using-agentmail"><img src="https://agentmods.dev/badge/skills/jtsternberg/claude-plugins/using-agentmail.svg" alt="Measured on agentmods" height="20"></a>
Per session 92 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 6,566 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00092 $0.06566
Opus 5 $0.00046 $0.03283
Sonnet 5 $0.00018 $0.01313
Haiku 4.5 $0.00009 $0.00657

Measured 3d ago against content hash be0de73c60de, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

using-agentmail scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/agentmail-signup.sh, scripts/agentmail-verify.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

> Then decide with the user. If a guaranteed-once send genuinely matters, drop to `curl`
plugins/agentmail/skills/using-agentmail/SKILL.md · 464 lines

How it starts

The opening of the file, as written. The whole thing — 464 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AgentMail

Give an agent its own email address and run a full round trip on it: create inboxes, send, read what comes back, reply / reply-all / forward, and stage drafts for review. Driven entirely by the official agentmail CLI.

This is AgentMail specifically — an email API built for agents, where every inbox is an API resource. It is not the user's own mail: reading or sending the user's Gmail is the gws plugin. It is not raw SMTP.

This is the hub. Four sibling skills go deeper

Use them instead of this file when the task is one of theirs; come back here for setup, flags, errors, and anything the others do not cover.

Skill For
contacts the address book — who is who, and which address is actually verified
check-mail what arrived, reading full bodies, and triaging an inbox
replying reply / reply-all / forward, draft-first for anything consequential
relay-work-order handing work to (or taking work from) another agent over email

Two shared references sit at the plugin root because more than one skill needs each: references/replying.md (the mechanics of answering mail) and references/agent-mail-protocol.md (the [HANDOFF]/[ASK]/[FYI]/[DONE] contract).

Codex: substitute the installed plugin directory for the path below.

${CLAUDE_PLUGIN_ROOT}/references/replying.md
${CLAUDE_PLUGIN_ROOT}/references/agent-mail-protocol.md

"What is your email address?" — check, don't assume

When you're asked whether you have an email, what your address is, or whether people can write to you, the honest answer is not "I'm an agent, I don't have one" and it is never the user's own address from session context. An inbox may have been provisioned for you already. Find out before you answer:

agentmail inboxes list

If it returns an inbox, that address (and its display name) is your answer. If it returns nothing — or the CLI/key isn't set up (run the preflight) — then say you have no inbox yet and offer to create one. Answering an identity question from memory is how an agent with a live @agentmail.to inbox came to tell its user, twice, that it had no email.

Read the full file on GitHub · 464 lines

Files

What ships with it

5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 464 lines · 92 tokens per session scan A be0de73c60de

Subscribe to this mod's changes

using-agentmail is a skill published in the GitHub repository jtsternberg/claude-plugins (5 stars, last pushed today), licensed MIT. It adds 92 tokens to every session and 6,566 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.