agentleFS
aashariaashari/ai-agent-skills/skills/apple-mail/mail-read/SKILL.md

mail-read

Skill7 starsChanged 9 months agoMITView on GitHub

Read the full content of a specific email by ROWID, subject search, or description. Shows headers, body, and any attachment names. Use when user wants to read a specific email, see the full text of a message, or asks "what does that email say". Arguments: ROWID number, subject keywords, or a description of the email to find.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If multiple matches, show options and ask which one.

    line 49
  2. Distinctive:

    If no summary, parse the emlx:

    line 73

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Mail Read — Read a Specific Email

Looking for: $ARGUMENTS

Step 1: Find the email

If $ARGUMENTS is a number → treat as ROWID directly

ROWID="$ARGUMENTS"
DB="$HOME/Library/Mail/V10/MailData/Envelope Index"

sqlite3 "$DB" "
SELECT datetime(m.date_received,'unixepoch','localtime') as dt,
       s.subject, a.address as sender, a.comment as name,
       mb.url as mailbox, m.ROWID, m.read, m.flagged
FROM messages m
JOIN subjects  s  ON m.subject = s.ROWID
JOIN addresses a  ON m.sender  = a.ROWID
JOIN mailboxes mb ON m.mailbox = mb.ROWID
WHERE m.ROWID = ${ROWID};" 2>/dev/null

If $ARGUMENTS is text → search by subject + sender

sqlite3 "$DB" "
SELECT datetime(m.date_received,'unixepoch','localtime') as dt,
       s.subject, a.address as sender, mb.url, m.ROWID
FROM messages m
JOIN subjects s ON m.subject=s.ROWID
JOIN addresses a ON m.sender=a.ROWID
JOIN mailboxes mb ON m.mailbox=mb.ROWID
WHERE (s.subject LIKE '%QUERY%' OR a.address LIKE '%QUERY%' OR a.comment LIKE '%QUERY%')
  AND m.deleted=0
  AND mb.url NOT LIKE '%Spam%'
ORDER BY m.date_received DESC LIMIT 5;" 2>/dev/null

If multiple matches, show options and ask which one.

Step 2: Get recipient list

sqlite3 "$DB" "
SELECT a.address, a.comment, r.type
FROM recipients r
JOIN addresses a ON r.address = a.ROWID
WHERE r.message = <ROWID>
ORDER BY r.type, r.position;" 2>/dev/null

Step 3: Get attachment list

sqlite3 "$DB" "
SELECT att.name FROM attachments att
WHERE att.message = <ROWID>;" 2>/dev/null

Step 4: Read the full body

First check summaries table for fast-path (no file I/O, available for ~5% of recent emails):

sqlite3 "$DB" "SELECT su.summary FROM messages m JOIN summaries su ON m.summary = su.ROWID WHERE m.ROWID = <ROWID>;" 2>/dev/null

If no summary, parse the emlx:

python3 ~/.claude/skills/_mail-shared/parser.py <ROWID>

Step 5: Find the raw file if needed

find ~/Library/Mail/V10/ -name "<ROWID>.emlx" 2>/dev/null | head -1
# If not found, try partial (IMAP lazy-load — body is still readable):
find ~/Library/Mail/V10/ -name "<ROWID>.partial.emlx" 2>/dev/null | head -1

The parser handles partial fallback automatically. Use raw find only for debugging.

Output Format

Email Details

From:     [name] <address>
To:       [recipients]
CC:       [if any]
Date:     [datetime]
Subject:  [subject]
Account:  [derived from mailbox URL]
Attachments: [filenames if any]

[Full body text, formatted naturally]


Offer:

  • "Want me to summarize this?" if long
  • "Want me to find related emails in this thread?"
  • "Should I help you draft a reply?"

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.