google-email

interact w/ Google Gmail inbox

설치로 이동

저장소: diegosouzapw/awesome-omni-skill
최근 소스 활동: 2026년 2월 28일 04:24
감지된 SKILL.md 언어: 영어
스타: 50
포크: 19

설치 방법

기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

직접 명령은 검토 Prompt를 거치지 않습니다. 실행하기 전에 소스를 확인하세요.

npx skills add https://github.com/diegosouzapw/awesome-omni-skill --skill google-email

명령은 한 줄로 유지됩니다. 복사하기 전에 가로로 스크롤해 전체 내용을 확인하세요.

로컬 사본을 원하시나요? SkillsMP에서 현재 제공할 수 있는 파일을 다운로드하세요.

소스 파일 검토

설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.

파일 탐색기

2 개 파일

SKILL.md

소스 지침 · 읽기 전용 미리보기

검색으로 돌아가기 설치로 돌아가기

artifact-tracking

Token-efficient tracking for AI orchestration. CLI-first for status updates (~50 tokens), agent fallback for complex ops (~1KB). Use when: updating task status, querying blockers, creating progress files, validating phases.

2026-03-02

ash-ai

AshAi extension guidelines for integrating AI capabilities with Ash Framework. Use when implementing vectorization/embeddings, exposing Ash actions as LLM tools, creating prompt-backed actions, or setting up MCP servers. Covers semantic search, LangChain integration, and structured outputs.

2026-03-02

ask-oracle

This skill should be used when solving hard questions, complex architectural problems, or debugging issues that benefit from GPT-5 Pro or GPT-5.1 thinking models with large file context. Use when standard Claude analysis needs deeper reasoning or extended context windows.

2026-03-02

이 저장소의 모든 Skills 보기

name	google-email
description	interact w/ Google Gmail inbox

Google Email

Overview

The google-email command is installed globally.

It is a dual-mode email management system for Gmail with offline analysis capabilities.

It provides:

Online Mode: Pull emails from Gmail to local Markdown storage
Offline Mode: Query and manage the stored email database without Gmail connectivity

The system is designed for AI agents to integrate email processing into workflows, supporting:

Batch email ingestion with deduplication
Email marking (read/unread) with offline metadata
Pattern analysis on stored emails
Integration with other tools and scripts

Core Concepts

Email IDs

Gmail ID: Alphanumeric identifier from Gmail API
SHA1 Hash: 40-character hex hash of Gmail ID, used as filename
Short ID: First 6 characters of hash (e.g., 6498ce), used for Git-like partial matching

Example:

Gmail ID:   18d676f08ff64932bf93e7ec3c0adb2b
SHA1 Hash:  6498cec18d676f0328ff649bf933e7ec3c0adb2b
Short ID:   6498ce

Storage Format

Each email is stored as a Markdown file in storage/ with YAML front matter containing metadata, and the HTML body in a code block:

---
id: '18d676f08ff64932bf93e7ec3c0adb2b'
threadId: '18d676f08ff64932'
subject: 'Project Status Update'
from:
  name: 'Alice Smith'
  address: 'asmith@company.com'
toRecipients:
  - name: 'Team'
    address: 'team@company.com'
receivedDateTime: '2026-01-05T10:30:00Z'
isRead: false
labelIds:
  - 'INBOX'
  - 'UNREAD'
webLink: 'https://mail.google.com/mail/u/0/#inbox/18d676f08ff64932bf93e7ec3c0adb2b'
body:
  contentType: html
_stored_id: '6498cec18d676f08ff64932bf93e7ec33c0adb2b'
_stored_at: '2026-01-05T18:05:13.476Z'
offline:
  read: true
  readAt: '2026-01-05T18:06:00.000Z'
---

# Project Status Update

```html
<html>
<head>...</head>
<body>
<p>Email content here...</p>
</body>
</html>


### ID Matching (Git-style)
All CLI commands support partial IDs. The system matches the longest unique prefix:

```bash
# These all refer to the same email:
google-email inbox view 6498cec18d676f08ff64932bf93e7ec33c0adb2b  # Full (40 chars)
google-email inbox view 6498cec18d676f08                          # 16 chars
google-email inbox view 6498ce                                     # 6 chars (short)
google-email inbox view 6498                                       # 4 chars (if unique)
google-email inbox view 6498cec18d676f08ff64932bf93e7ec33c0adb2b.md  # Filename format

Error on ambiguity:

Error: Ambiguous ID "62". Matches: 62e8e2d5adb20b15..., 62b19cb17ec4628a...

Online Mode: Fetching Emails from Gmail

Command: `google-email pull`

Purpose: Fetch unread emails from Gmail, store locally as Markdown files, mark as read/processed in Gmail.

IMPORTANT: When instructed to fetch more emails, use this command. Always fetch exactly one email at a time (--limit 1) unless specifically directed to pull more.

Command:

google-email pull --since <date> [--limit N]

Parameters:

--since <date>: Required. Fetch emails received on/after this date
- Formats: YYYY-MM-DD, yesterday, "7 days ago", "1 day ago"
--limit <n>: Optional. Stop after processing N emails (default: no limit)

Behavior:

Fetches all unread emails from inbox since date
Paginates through results
Skips already-stored files (deduplication via SHA1)
Stores each new email as Markdown under storage/<id>.md
Marks processed emails as read in Gmail
Adds "Processed" label and archives emails

Examples:

# RECOMMENDED: Pull exactly 1 new email (safest, most controlled)
google-email pull --since yesterday --limit 1

# Pull from specific date, one at a time
google-email pull --since 2026-01-01 --limit 1

# Pull from last week, one at a time
google-email pull --since "7 days ago" --limit 1

# Pull multiple emails (only when specifically instructed)
google-email pull --since yesterday --limit 5

Output Example:

Fetching unread emails since: 2026-01-05
Processing limit: 1
Found 14 unread emails.
✓ Stored: (71c95a+NOTICE: Password Expiration)
  → Marking as read...
  → Adding Processed label and archiving...
  ✓ Updated in Gmail

Reached processing limit of 1. Stopping.

Summary:
  Available:  14
  Processed:  1
  Written:    1
  Skipped:    0

Offline Mode: Analysis & Metadata

Local Query & Management

All google-email commands work purely offline, reading/writing Markdown files in storage/.

Command: `google-email inbox summary`

Purpose: Get overall email statistics

Command:

google-email inbox summary

Output:

Folder Summary:
===============
Inbox:
  Unread: 46
  Read:   1
  Total:  47

Overall:
  Unread: 46
  Read:   1
  Total:  47

Command: `google-email inbox list [OPTIONS]`

Purpose: List emails with filtering and display

Command:

google-email inbox list [--limit N] [--since DATE] [--all]

Options:

-l, --limit <n>: Max results (default: 10)
--since <date>: Filter emails after date
-a, --all: Include read emails (default: unread only)

Output Example:

Showing 3 of 46 emails:

f86bca / Today 10:00 / Science Operations <ScienceOperations@company.com>
Science record review request

cda64a / Today 9:46 / Alice Johnson <alice@company.com>
Re: [team] Create new feature (PR #537)

8fb5bf / Today 9:01 / U.S. Payroll <payroll@company.com>
Payroll for 2026

... and 43 more

Command: `google-email inbox view <id>`

Purpose: Display full YAML content of an email

Command:

google-email inbox view <id>

Supports: Partial IDs, full IDs, or filename format

Command: `google-email inbox read <id>`

Purpose: Mark an email as read (offline only)

Command:

google-email inbox read <id>

Effect:

Adds/updates offline.read: true to YAML front matter
Adds offline.readAt timestamp
Does NOT sync back to Gmail (offline only)

Command: `google-email inbox unread <id>`

Purpose: Mark an email as unread (offline only, reverses read state)

Command:

google-email inbox unread <id>

Integration Patterns for AI Agents

Pattern 1: Fetch & Analyze

# Typical agent workflow
1. google-email pull --since yesterday       # Get new emails
2. google-email inbox summary                # Get stats
3. google-email inbox list --limit 50        # Scan subjects
4. google-email inbox view <id>              # Get full content
5. <agent analyzes>
6. google-email inbox read <id>              # Mark processed

Pattern 2: Incremental Processing

# Process in batches to avoid overwhelming
1. google-email pull --since yesterday --limit 20    # Get batch
2. for each email:
   a. google-email inbox view <id> | extract content
   b. Send to AI for analysis
   c. Store result in database
   d. google-email inbox read <id>  # Mark done
3. Repeat if more emails available

Data Access Examples

Extract Sender Name

google-email inbox view 6498ce | yq '.from.name'
# Output: "Alice Smith"

Extract All Recipients

google-email inbox view 6498ce | yq '.toRecipients[].address'

Extract Email Subject

google-email inbox view 6498ce | yq '.subject'

Extract Received Date

google-email inbox view 6498ce | yq '.receivedDateTime'

Check Read Status (Offline)

google-email inbox view 6498ce | yq '.offline.read'

Tips for AI Agents

Always check summary first: google-email inbox summary
Use partial IDs: Shorter 6498ce instead of full hash
Pipe to tools: google-email inbox view <id> | yq '.field'
Test with limit: Use --limit 5 or --limit 1 when prototyping
Mark processed: Always read after processing to track state
Use relative dates: yesterday, "7 days ago" are clearer

Architecture Notes

Storage: All files in storage/ are Markdown with YAML front matter (Git-friendly)
No database: Direct filesystem access via Bun file API
Stateless: Each command is independent
Deduplication: SHA1 hashing ensures same Gmail ID = same storage file
Offline metadata: offline.* fields are never overwritten by pull
Single direction: pull syncs Gmail → storage; read/unread mark locally only

google-email

소스 정보

설치 방법

소스 파일 검토

이 저장소의 다른 Skills

Google Email

Overview

Core Concepts

Email IDs

Storage Format

Online Mode: Fetching Emails from Gmail

Command: google-email pull

Offline Mode: Analysis & Metadata

Local Query & Management

Command: google-email inbox summary

Command: google-email inbox list [OPTIONS]

Command: google-email inbox view <id>

Command: google-email inbox read <id>

Command: google-email inbox unread <id>

Integration Patterns for AI Agents

Pattern 1: Fetch & Analyze

Pattern 2: Incremental Processing

Data Access Examples

Extract Sender Name

Extract All Recipients

Extract Email Subject

Extract Received Date

Check Read Status (Offline)

Tips for AI Agents

Architecture Notes

이 저장소의 다른 Skills

Google Email

Overview

Core Concepts

Email IDs

Storage Format

Online Mode: Fetching Emails from Gmail

Command: google-email pull

Offline Mode: Analysis & Metadata

Local Query & Management

Command: google-email inbox summary

Command: google-email inbox list [OPTIONS]

Command: google-email inbox view <id>

Command: google-email inbox read <id>

Command: google-email inbox unread <id>

Integration Patterns for AI Agents

Pattern 1: Fetch & Analyze

Pattern 2: Incremental Processing

Data Access Examples

Extract Sender Name

Extract All Recipients

Extract Email Subject

Extract Received Date

Check Read Status (Offline)

Tips for AI Agents

Architecture Notes

Command: `google-email pull`

Command: `google-email inbox summary`

Command: `google-email inbox list [OPTIONS]`

Command: `google-email inbox view <id>`

Command: `google-email inbox read <id>`

Command: `google-email inbox unread <id>`

Command: `google-email pull`

Command: `google-email inbox summary`

Command: `google-email inbox list [OPTIONS]`

Command: `google-email inbox view <id>`

Command: `google-email inbox read <id>`

Command: `google-email inbox unread <id>`