gmail-automation
Automate Gmail tasks via Rube MCP (Composio): send/reply, search, labels, drafts, attachments. Always search tools first for current schemas.
Install with Tessl CLI
npx tessl i github:davepoon/buildwithclaude --skill gmail-automation72
Quality
60%
Does it follow best practices?
Impact
96%
1.52xAverage score across 3 eval scenarios
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/all-skills/skills/gmail-automation/SKILL.mdEvaluation results
92%
8%Inbox Triage Automation for a Busy Engineering Team
Search query syntax and bulk label operations
RUBE_SEARCH_TOOLS first
0%
0%
LIST_LABELS before create
100%
100%
is:unread syntax
100%
100%
Date range query
0%
100%
Label ID not name
100%
100%
BATCH_MODIFY for bulk
100%
100%
REMOVE UNREAD to read
100%
100%
Pagination handling
100%
100%
Batch size limit
100%
100%
No include_payload on search
100%
100%
Without context: $0.2131 · 1m 18s · 12 turns · 19 in / 4,125 out tokens
With context: $0.3779 · 1m 36s · 16 turns · 268 in / 5,217 out tokens
100%
65%Customer Support Email Response Automation
Thread reply workflow and ID handling
RUBE_SEARCH_TOOLS first
0%
100%
FETCH_EMAILS prerequisite
80%
100%
REPLY_TO_THREAD not SEND_EMAIL
0%
100%
No subject in reply
0%
100%
Hex thread_id from results
80%
100%
No legacy web UI IDs
0%
100%
is: syntax for search
100%
100%
is_html for HTML replies
20%
100%
Defensive response parsing
0%
100%
from: domain search syntax
100%
100%
Without context: $0.2802 · 1m 37s · 14 turns · 69 in / 5,201 out tokens
With context: $0.2642 · 1m 12s · 13 turns · 264 in / 3,701 out tokens
97%
25%Stakeholder Outreach Draft Preparation
Draft creation with attachments and contact resolution
RUBE_SEARCH_TOOLS first
0%
62%
SEARCH_PEOPLE for names
41%
100%
is_html for HTML body
25%
100%
Mimetype includes slash
100%
100%
data.id for draft ops
100%
100%
UPDATE_DRAFT includes all fields
100%
100%
Exponential backoff
100%
100%
Explicit approval before send
100%
100%
LIST_DRAFTS for verification
100%
100%
s3key expiry note
0%
100%
Without context: $0.3626 · 2m 1s · 17 turns · 1,663 in / 6,368 out tokens
With context: $0.2429 · 1m 8s · 11 turns · 263 in / 3,756 out tokens
- 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.