email-triage
Triage Joel's email inboxes via the joelclaw email CLI. Scan, categorize, archive noise, surface actionable items, and draft replies. Use when: 'check my email', 'scan inbox', 'triage email', 'what needs a reply', 'clean up inbox', 'archive junk', 'email summary', 'anything important in email', or any request involving email inbox review or cleanup.
90
88%
Does it follow best practices?
Impact
92%
2.87xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Evaluation results
95%
67%Daily Email Triage Script
Triage workflow and summary format
Uses joelclaw email CLI
50%
100%
is:unreplied not is:unread
0%
100%
archive-ids for batch noise
0%
100%
Four triage categories
25%
100%
Summary section order
25%
100%
VIP thread protection
16%
100%
Draft-then-approve for replies
91%
100%
Inference-based categorization
10%
50%
Inbox scan command
30%
100%
Read conversation command
30%
100%
82%
51%Front Email Search Query Reference Script
Front search query construction
is:unreplied not is:unread
0%
100%
Unix timestamps in API queries
50%
100%
CLI date flag convenience
0%
100%
No negation operators
40%
100%
Same-type filter OR logic
87%
100%
Different-type filter AND logic
87%
100%
Entity filters use IDs
50%
100%
Mutually exclusive statuses
37%
100%
Max 15 filters limit
0%
100%
Backwards-compatible body extraction
0%
0%
Cache bypass option
0%
0%
100%
63%Inbox Bulk Cleanup Script
Bulk archive cleanup with safety checks
archive-bulk dry run first
0%
100%
archive-bulk --confirm to execute
0%
100%
github.com notifications cleanup
100%
100%
betterstack alerts cleanup
0%
100%
Stale CI notification handling
50%
100%
Duplicate conversation handling
100%
100%
Rate limit awareness
60%
100%
archive-ids for page results
0%
100%
Uses is:open status filter
0%
100%
Pagination handling
0%
100%
No negation in queries
100%
100%
- Repository
- joelhooks/joelclaw
- Commit
825972c
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.