A secure, extraction-only skill for processing forwarded emails through a dedicated IMAP inbox with zero automatic background activity.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install safe-email
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 safe-email 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 Safe Email?
Safe Email is a privacy-centric component of Openclaw Skills designed to bridge the gap between your inbox and AI agents without compromising security. By utilizing a dedicated inbox and requiring explicit user triggers, it ensures that your sensitive data is never subject to automated background polling or unauthorized external writes.
This skill focuses exclusively on extraction, converting messy email content into structured, actionable data. It provides a sandboxed environment where forwarded emails are parsed for key information like dates, locations, and confirmation numbers, allowing you to maintain full control over what information is processed and how it is used in subsequent workflows.
Safe Email Use Cases
- Extracting meeting details or travel itineraries from forwarded confirmation emails.
- Processing specific newsletters or alerts for structured summary generation.
- Converting forwarded tasks into a clean list of actionable items for manual review.
- Sanitizing email data before manually passing it to other productivity tools within the Openclaw Skills ecosystem.
How Safe Email Works
- The skill waits for an explicit user command to process the latest forwarded email.
- It connects to the dedicated IMAP inbox using the Himalaya CLI and retrieves only the newest relevant message.
- The engine parses the email body to identify structured details such as senders, timestamps, links, and notes.
- A structured summary is presented to the user along with suggested next steps for manual confirmation.
- If requested, the email is moved to the trash or expunged only after explicit user consent is granted.
Safe Email Setup
To begin using this skill, ensure you have a dedicated email account with IMAP enabled and the Himalaya CLI installed. This setup is optimized for the Openclaw Skills framework.
- Install the Himalaya CLI:
# On macOS using Homebrew
brew install himalaya
- Configure your environment variables for secure access:
export SAFE_EMAIL_IMAP_USERNAME="[email protected]"
export SAFE_EMAIL_IMAP_APP_PASSWORD="your-app-specific-password"
- Verify your IMAP settings are pointed to
imap.gmail.com:993with TLS enabled.
Safe Email Data Schema & Taxonomy
Safe Email organizes extracted data into a clean, predictable schema to ensure compatibility with other Openclaw Skills. The following metadata taxonomy is used:
| Field | Description |
|---|---|
| sender | The original sender of the forwarded message |
| subject | The email subject or title line |
| date_time | Extracted time window, including timezone if available |
| location | Physical or virtual meeting locations identified in the text |
| links | All URLs and actionable links discovered in the body |
| key_notes | Confirmation numbers, seat assignments, or participant lists |
| actionable_items | A synthesized list of suggested next steps for the user |
Safe Email Advanced Features
- Uses the Himalaya CLI for lightweight and robust IMAP interactions.
- Extraction-only architecture prevents unauthorized writes to external systems like calendars or reminders.
- Strict compliance with privacy protocols, including the forbidden use of auto-polling or background surveillance.
- Destructive action protection that requires explicit multi-step consent for email deletion.
- Modular design allows the extracted data to be piped into other Openclaw Skills for further automation.
SKILL.md
Loading
Related Openclaw Skills
Vector Store Shootout
A unified benchmarking and implementation suite for eight leading vector database backends using a common interface.
Tweet Pipeline
A Notion-to-Twitter automation bridge that schedules and posts tweets directly from a content calendar.
Smoke Test Generator
A comprehensive tool for generating and running categorized API smoke test suites to ensure deployment readiness and service health.
LLM Judge Ensemble
A multi-layered evaluation framework for scoring AI outputs using deterministic checks, heuristics, and sampled LLM judges.
Nano Banana Cog
A sophisticated image generation skill that brings Google DeepMind's Nano Banana model to AI agents for photorealistic creation and character consistency.
Sheet Cog
An advanced AI coding agent skill that generates complex, formula-heavy spreadsheets and financial models using Python-powered automation.
METADATA
Featured*
