A secure, read-only skill for parsing and executing mailbox retrieval commands against a local Gmail RAG backend.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install clawinboxrag
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install clawinboxrag using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is ClawInboxRAG?
ClawInboxRAG is a specialized community extension within the Openclaw Skills ecosystem designed to bridge the gap between AI agents and local mail repositories. It enables users to interact with their Gmail data through a structured command-line interface, utilizing Python-based parsing to execute safe, read-only queries against a local backend while maintaining strict security boundaries.
By leveraging localized scripts, this skill allows for advanced retrieval augmented generation (RAG) workflows. It ensures that sensitive email data is accessed only through validated actions, providing concise and citation-friendly outputs that are perfect for high-context AI interactions.
ClawInboxRAG Use Cases
- Retrieving recent emails using simple natural language commands like mail recents.
- Performing hybrid searches that combine keyword and semantic logic across local mail archives.
- Syncing local mail databases with Gmail servers while maintaining a read-only posture.
- Filtering specific messages by labels or date ranges to provide context for complex research tasks.
How ClawInboxRAG Works
- The skill detects chat input starting with the trigger word mail.
- The internal parser analyzes the command to identify actions such as search, sync, status, or help.
- Parameters like search mode (semantic, keyword, or hybrid) and result limits are extracted and validated against safety constraints.
- Validated requests are routed to a shell script that interfaces with the local Gmail RAG repository.
- The skill returns a bounded, formatted response to the agent, ensuring no sensitive credentials or excessive data volumes are exposed.
ClawInboxRAG Setup
To integrate this skill into your Openclaw Skills environment, ensure you have a local backend checkout and the uv runner installed.
- Configure your Gmail OAuth read-only scope credentials.
- Set the necessary environment variables:
export GMAIL_RAG_REPO="/path/to/your/repo"
export MAIL_DEFAULT_MODE="hybrid"
export MAIL_MAX_LIMIT=25
- Ensure the scripts/parse_mail.py and scripts/run_cli.sh files are executable within your environment.
ClawInboxRAG Data Schema & Taxonomy
The skill organizes mailbox data through a structured parsing logic and interacts with the backend using the following schema:
| Action | Parameters | Description |
|---|---|---|
| search | query, mode, limit, date filters | Executes retrieval using keyword, semantic, or hybrid methods. |
| sync | ingest, embed, refresh | A multi-step workflow to update local mail indexes. |
| recents | limit | Returns the most recent messages up to the configured maximum. |
| status | N/A | Provides current state of the local mailbox backend. |
ClawInboxRAG Advanced Features
- Support for localized aliases and multilingual summary flags like résumé.
- Automated sync workflows that handle ingestion and embedding in a single logical operation.
- Granular date filtering (after, before, between) for precise historical data retrieval.
- Strict safety posture that enforces numeric limits on search results to prevent token overflow.
SKILL.md
Loading
Related Openclaw Skills
Lottery Predictor V3.6
A sophisticated data-driven lottery prediction engine for Openclaw Skills utilizing an 8-dimensional intelligent scoring system and historical backtesting.
gen-design
An AI-powered system design assistant that generates architecture documents and data models through interactive technology selection and domain knowledge integration.
gen-coding-specs
A powerful documentation engine that generates a complete suite of technical coding specifications and style guides to align AI agents and human developers.
Doubao Podcast TTS Integration
A specialized integration skill for the ByteDance Doubao Podcast TTS API, enabling high-quality audio generation from articles and text via a custom binary WebSocket protocol.
Incident Response Playbook Generator API
Automate the creation of tailored incident response playbooks based on organizational data and security compliance requirements.
Shell Image & Video Skill
A powerful AI media generation toolkit for face swapping, motion transfer, and digital human creation via RunningHub AI.
Featured*
