mail-top-senders
Skill7 starsChanged 9 months agoMITView on GitHub
Show who sends the most email, communication frequency analysis, and relationship mapping. Use when user asks who emails them most, top contacts, communication patterns, or wants to understand their email social graph. Arguments: optional time range (default: last 90 days), account filter, or "humans only" to exclude automated senders.
In this repository32 other files it gives its agents.
- calendar-accountsSkill
- calendar-attendeesSkill
- calendar-conflictsSkill
- calendar-coreSkill
- calendar-eventSkill
- calendar-freeSkill
- calendar-recurringSkill
- calendar-searchSkill
- calendar-statsSkill
- calendar-todaySkill
- calendar-upcomingSkill
- mail-accountsSkill
- mail-action-itemsSkill
- mail-attachmentsSkill
- mail-bankingSkill
- mail-contactsSkill
- mail-coreSkill
- mail-digestSkill
- mail-expensesSkill
- mail-fromSkill
- mail-meetingsSkill
- mail-needs-replySkill
- mail-newsletterSkill
- mail-readSkill
- mail-searchSkill
- mail-securitySkill
- mail-statsSkill
- mail-subscriptionsSkill
- mail-threadSkill
- mail-travelSkill
- mail-triageSkill
- mail-workSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 42
Use automated_conversation and unsubscribe_type columns (more reliable than address-pattern matching):
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.
Files
Mail Top Senders — Communication Frequency Analysis
Analysis period: $ARGUMENTS (default: last 90 days)
Step 1: All senders ranked by volume
DB="$HOME/Library/Mail/V10/MailData/Envelope Index"
SINCE=$(($(date +%s) - 7776000)) # 90 days
sqlite3 "$DB" "
SELECT a.address, a.comment as name,
COUNT(*) as total,
SUM(CASE WHEN m.read=0 THEN 1 ELSE 0 END) as unread,
MIN(datetime(m.date_received,'unixepoch','localtime')) as first,
MAX(datetime(m.date_received,'unixepoch','localtime')) as latest
FROM messages m
JOIN addresses a ON m.sender = a.ROWID
JOIN mailboxes mb ON m.mailbox = mb.ROWID
WHERE m.date_received >= ${SINCE}
AND m.deleted = 0
AND mb.url NOT LIKE '%Spam%' AND mb.url NOT LIKE '%Trash%'
AND mb.url NOT LIKE '%Sent%'
GROUP BY a.address
ORDER BY total DESC
LIMIT 50;" 2>/dev/nullStep 2: Separate humans from automated senders
Use automated_conversation and unsubscribe_type columns (more reliable than address-pattern matching):
automated_conversation = 0+unsubscribe_type = 0→ real humansautomated_conversation = 1→ transactional (Jira, Slack, alerts)automated_conversation = 2ORunsubscribe_type > 0→ bulk/newsletters (noise)
# Human senders only (automated_conversation = 0, no unsubscribe header)
sqlite3 "$DB" "
SELECT a.address, a.comment as name, COUNT(*) as cnt
FROM messages m
JOIN addresses a ON m.sender = a.ROWID
JOIN mailboxes mb ON m.mailbox = mb.ROWID
WHERE m.date_received >= ${SINCE}
AND m.deleted = 0
AND mb.url NOT LIKE '%Spam%' AND mb.url NOT LIKE '%Sent%'
AND m.automated_conversation = 0
AND m.unsubscribe_type = 0
GROUP BY a.address ORDER BY cnt DESC LIMIT 20;" 2>/dev/nullStep 3: Domain-level analysis
sqlite3 "$DB" "
SELECT substr(a.address, instr(a.address,'@')+1) as domain,
COUNT(*) as cnt
FROM messages m
JOIN addresses a ON m.sender = a.ROWID
JOIN mailboxes mb ON m.mailbox = mb.ROWID
WHERE m.date_received >= ${SINCE}
AND m.deleted = 0
AND mb.url NOT LIKE '%Spam%'
GROUP BY domain ORDER BY cnt DESC LIMIT 20;" 2>/dev/nullStep 4: Thread participation (two-way communication)
Find senders you also replied to — true relationships vs. one-way communication:
sqlite3 "$DB" "
SELECT a.address, COUNT(*) as received
FROM messages m
JOIN addresses a ON m.sender = a.ROWID
JOIN mailboxes mb ON m.mailbox = mb.ROWID
WHERE m.date_received >= ${SINCE} AND m.deleted=0
AND mb.url NOT LIKE '%Sent%'
GROUP BY a.address
ORDER BY received DESC LIMIT 30;" 2>/dev/nullOutput Format
Top Email Relationships — [PERIOD]
Most Frequent Human Contacts:
| Rank | Name | Address | Emails Received | Unread |
Top Automated Senders (noise):
| Service | Count | Type |
Top Domains:
| Domain | Count |
Insight: note anyone with high unread rate (you receive a lot but don't read → de-prioritize subscription?). Note anyone with very recent last email who you haven't read → potential missed message.
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
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.

