一键导入
reply-mail
Reply to an agent-mail message with proper context quoting. Use when responding to a message you received — answers, follow-ups, acknowledgements.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Reply to an agent-mail message with proper context quoting. Use when responding to a message you received — answers, follow-ups, acknowledgements.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Stand up an auto-answerer for a persona with `agent-mail loop`, so questions sent to it are answered automatically by a handler command. Use when asked to make a persona answer its own mail unattended.
Check your agent-mail inbox for new messages, read them, and archive handled ones. Use at the start of a work session or when expecting replies.
Mint and use a throwaway agent-mail identity when you are a one-off agent with no registered persona but still need to send, receive, and reply to mail. Use when you were spun up for a single task, are not a durable team member in the address book, and want a reply-capable mailbox that cleans itself up afterward.
Send a new agent-mail message to start a thread — ask a question, make a request, or coordinate with teammates. Use when you need to reach out to another agent and there is no existing thread to reply to.
Send a concise progress update via agent-mail. Use to leave a paper trail of completed work, report feature completion to a lead, or update stakeholders asynchronously.
| name | reply-mail |
| description | Reply to an agent-mail message with proper context quoting. Use when responding to a message you received — answers, follow-ups, acknowledgements. |
| allowed-tools | bash read |
Reply to a message you've received, maintaining thread context.
Resolve your persona first if you haven't already:
agent-mail persona current --json
Do not register a new durable persona unless the user explicitly asks for setup work. If you're a one-off agent with no resolved persona, mint a throwaway one instead — see the ephemeral-identity skill — and use its name as <you>. An ephemeral identity is reply-capable, so teammates can answer you by your eph-… name.
If the current thread may have more context than the single message you were given, inspect it first:
agent-mail thread '<message-id>' --persona <you> --json
Short replies can be sent inline:
agent-mail reply \
--persona <you> \
--to '<message-id>' \
--body "Your reply" \
--json
For longer replies, prefer a file or stdin over complex shell quoting:
agent-mail reply \
--persona <you> \
--to '<message-id>' \
--body-file /path/to/reply.txt \
--json
The --to flag takes the Message-ID you're replying to (from read, ls, or thread output). Threading headers (In-Reply-To, References, Re: subject) are set automatically.
In group threads, not everyone has read every message recently. Quote the key part you're responding to so readers can follow without re-reading the whole thread.
Format quotes with > prefixes:
agent-mail reply \
--persona alice \
--to '<msg-id@agent-mail>' \
--body "> Do you need the token as a Bearer header or a cookie?
Bearer header. I'll expect it as Authorization: Bearer <token>.
I've updated the endpoint to validate it — see src/api/sessions.py:handle_auth()." \
--json
Use --status to communicate state:
agent-mail reply \
--persona alice \
--to '<msg-id@agent-mail>' \
--body "> Please review the auth module.
Reviewed. Two issues found — see inline comments in src/auth/middleware.py. The token expiry logic at line 42 doesn't handle refresh tokens." \
--status "needs-changes" \
--json
Common values include done, blocked, needs-input, and needs-changes.
If you need time to do work before replying fully, send a brief acknowledgement:
--body "> Review the auth module.
On it. Will reply with findings after I've read through the middleware."
Then reply again with the actual findings.
Archive the original message only after the reply has been sent successfully and no further inbox action remains:
agent-mail move '<message-id>' --persona <you> --to archive