gmail-automation
Automate Gmail tasks via Rube MCP (Composio): send/reply, search, labels, drafts, attachments. Always search tools first for current schemas.
What this skill does
# Gmail Automation via Rube MCP
Automate Gmail operations through Composio's Gmail toolkit via Rube MCP.
## Prerequisites
- Rube MCP must be connected (RUBE_SEARCH_TOOLS available)
- Active Gmail connection via `RUBE_MANAGE_CONNECTIONS` with toolkit `gmail`
- Always call `RUBE_SEARCH_TOOLS` first to get current tool schemas
## Setup
**Get Rube MCP**: Add `https://rube.app/mcp` as an MCP server in your client configuration. No API keys needed — just add the endpoint and it works.
1. Verify Rube MCP is available by confirming `RUBE_SEARCH_TOOLS` responds
2. Call `RUBE_MANAGE_CONNECTIONS` with toolkit `gmail`
3. If connection is not ACTIVE, follow the returned auth link to complete Google OAuth
4. Confirm connection status shows ACTIVE before running any workflows
## Core Workflows
### 1. Send an Email
**When to use**: User wants to compose and send a new email
**Tool sequence**:
1. `GMAIL_SEARCH_PEOPLE` - Resolve contact name to email address [Optional]
2. `GMAIL_SEND_EMAIL` - Send the email [Required]
**Key parameters**:
- `recipient_email`: Email address or 'me' for self
- `subject`: Email subject line
- `body`: Email content (plain text or HTML)
- `is_html`: Must be `true` if body contains HTML markup
- `cc`/`bcc`: Arrays of email addresses
- `attachment`: Object with `{s3key, mimetype, name}` from prior download
**Pitfalls**:
- At least one of `recipient_email`, `cc`, or `bcc` required
- At least one of `subject` or `body` required
- Attachment `mimetype` MUST contain '/' (e.g., 'application/pdf', not 'pdf')
- Total message size limit ~25MB after base64 encoding
- Use `from_email` only for verified aliases in Gmail 'Send mail as' settings
### 2. Reply to a Thread
**When to use**: User wants to reply to an existing email conversation
**Tool sequence**:
1. `GMAIL_FETCH_EMAILS` - Find the email/thread to reply to [Prerequisite]
2. `GMAIL_REPLY_TO_THREAD` - Send reply within the thread [Required]
**Key parameters**:
- `thread_id`: Hex string from FETCH_EMAILS (e.g., '169eefc8138e68ca')
- `message_body`: Reply content
- `recipient_email`: Reply recipient
- `is_html`: Set `true` for HTML content
**Pitfalls**:
- `thread_id` must be hex string; prefixes like 'msg-f:' are auto-stripped
- Legacy Gmail web UI IDs (e.g., 'FMfcgz...') are NOT supported
- Subject is inherited from original thread; setting it creates a new thread instead
- Do NOT include subject parameter to stay within thread
### 3. Search and Filter Emails
**When to use**: User wants to find specific emails by sender, subject, date, label, etc.
**Tool sequence**:
1. `GMAIL_FETCH_EMAILS` - Search with Gmail query syntax [Required]
2. `GMAIL_FETCH_MESSAGE_BY_MESSAGE_ID` - Get full message details for selected results [Optional]
**Key parameters**:
- `query`: Gmail search syntax (from:, to:, subject:, is:unread, has:attachment, after:YYYY/MM/DD, before:YYYY/MM/DD)
- `max_results`: 1-500 messages per page
- `label_ids`: System IDs like 'INBOX', 'UNREAD'
- `include_payload`: Set `true` to get full message content
- `ids_only`: Set `true` for just message IDs
- `page_token`: For pagination (from `nextPageToken`)
**Pitfalls**:
- Returns max ~500 per page; follow `nextPageToken` via `page_token` until absent
- `resultSizeEstimate` is approximate, not exact count
- Use 'is:' for states (is:unread, is:snoozed, is:starred)
- Use 'label:' ONLY for user-created labels
- Common mistake: 'label:snoozed' is WRONG — use 'is:snoozed'
- `include_payload=true` on broad searches creates huge responses; default to metadata
- Custom labels require label ID (e.g., 'Label_123'), NOT label name
### 4. Manage Labels
**When to use**: User wants to create, modify, or organize labels
**Tool sequence**:
1. `GMAIL_LIST_LABELS` - List all labels to find IDs and detect conflicts [Required]
2. `GMAIL_CREATE_LABEL` - Create a new label [Optional]
3. `GMAIL_PATCH_LABEL` - Rename or change label colors/visibility [Optional]
4. `GMAIL_DELETE_LABEL` - Delete a user-created label (irreversible) [Optional]
**Key parameters**:
- `label_name`: Max 225 chars, no commas, '/' for nesting (e.g., 'Work/Projects')
- `background_color`/`text_color`: Hex values from Gmail's predefined palette
- `id`: Label ID for PATCH/DELETE operations
**Pitfalls**:
- 400/409 error if name is blank, duplicate, or reserved (INBOX, SPAM, CATEGORY_*)
- Color specs must use Gmail's predefined palette of 102 hex values
- DELETE is permanent and removes label from all messages
- Cannot delete system labels (INBOX, SENT, DRAFT, etc.)
### 5. Apply/Remove Labels on Messages
**When to use**: User wants to label, archive, or mark emails as read/unread
**Tool sequence**:
1. `GMAIL_LIST_LABELS` - Get label IDs for custom labels [Prerequisite]
2. `GMAIL_FETCH_EMAILS` - Find target messages [Prerequisite]
3. `GMAIL_BATCH_MODIFY_MESSAGES` - Bulk add/remove labels (up to 1000 messages) [Required]
4. `GMAIL_ADD_LABEL_TO_EMAIL` - Single-message label changes [Fallback]
**Key parameters**:
- `messageIds`: Array of message IDs (max 1000)
- `addLabelIds`: Array of label IDs to add
- `removeLabelIds`: Array of label IDs to remove
- `message_id`: 15-16 char hex string for single operations
**Pitfalls**:
- Max 1000 messageIds per BATCH call; chunk larger sets
- Use 'CATEGORY_UPDATES' not 'UPDATES'; full prefix required for category labels
- SENT, DRAFT, CHAT are immutable — cannot be added/removed
- To mark as read: REMOVE 'UNREAD'. To archive: REMOVE 'INBOX'
- `message_id` must be 15-16 char hex, NOT UUIDs or web UI IDs
### 6. Handle Drafts and Attachments
**When to use**: User wants to create, edit, or send email drafts, possibly with attachments
**Tool sequence**:
1. `GMAIL_CREATE_EMAIL_DRAFT` - Create a new draft [Required]
2. `GMAIL_UPDATE_DRAFT` - Edit draft content [Optional]
3. `GMAIL_LIST_DRAFTS` - List existing drafts [Optional]
4. `GMAIL_SEND_DRAFT` - Send a draft (requires explicit user approval) [Optional]
5. `GMAIL_GET_ATTACHMENT` - Download attachment from existing message [Optional]
**Key parameters**:
- `recipient_email`: Draft recipient
- `subject`: Draft subject (omit for reply drafts to stay in thread)
- `body`: Draft content
- `is_html`: Set `true` for HTML content
- `attachment`: Object with `{s3key, mimetype, name}`
- `thread_id`: For reply drafts (leave subject empty to stay in thread)
**Pitfalls**:
- Response includes `data.id` (draft_id) AND `data.message.id`; use `data.id` for draft operations
- Setting subject on a thread reply draft creates a NEW thread instead
- Attachment capped at ~25MB; base64 overhead can push near-limit files over
- UPDATE_DRAFT replaces entire content, not patches; include all fields you want to keep
- HTTP 429 on bulk draft creation; use exponential backoff
## Common Patterns
### ID Resolution
**Label name → Label ID**:
```
1. Call GMAIL_LIST_LABELS
2. Find label by name in response
3. Extract id field (e.g., 'Label_123')
```
**Contact name → Email**:
```
1. Call GMAIL_SEARCH_PEOPLE with query=contact_name
2. Extract emailAddresses from response
```
**Thread ID from search**:
```
1. Call GMAIL_FETCH_EMAILS or GMAIL_LIST_THREADS
2. Extract threadId (15-16 char hex string)
```
### Pagination
- Set `max_results` up to 500 per page
- Check response for `nextPageToken`
- Pass token as `page_token` in next request
- Continue until `nextPageToken` is absent or empty string
- `resultSizeEstimate` is approximate, not exact
### Gmail Query Syntax
**Operators**:
- `from:[email protected]` - Emails from sender
- `to:[email protected]` - Emails to recipient
- `subject:"exact phrase"` - Subject contains exact phrase
- `is:unread` - Unread messages
- `is:starred` - Starred messages
- `is:snoozed` - Snoozed messages
- `has:attachment` - Has attachments
- `after:2024/01/01` - After date (YYYY/MM/DD)
- `before:2024/12/31` - Before date
- `label:custom_label` - User-created label (use label ID)
- `in:sent` - In sent folder
- `category:primary` - Primary category
**Combinators**:
- `AND` - Both coRelated in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.