wecomcli-email

Send, reply, forward, search, and read WeCom emails via the wecom-cli mail interface.

2.0k|291|Updated Jul 24, 2026
One-click install
npx skills add https://github.com/Pinvou/pinvou-agent --skill wecomcli-email-pinvou
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wecomcli-email
Source: https://github.com/Pinvou/pinvou-agent/tree/main/pinvou3-app/src-tauri/resources/common/bundle/wecom-skills/wecomcli-email
Command: npx skills add https://github.com/Pinvou/pinvou-agent --skill wecomcli-email-pinvou

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires wecom-cli, and includes references (resource) components.

What problem does it solve? Managing a WeCom (Enterprise WeChat) mailbox through natural language is hard: recipients must be resolved from the contact directory, bodies must be written as Markdown files, attachments and inline images need media handling, and reply/forward subjects need correct prefix construction. This Skill orchestrates the entire email workflow through the wecom-cli mail commands so users can handle their corporate mailbox conversationally. ## Core Features & Use Cases - Send, reply, and forward emails: Compose Markdown bodies, attach files or inline images, resolve recipients by name via the contact directory, and preview the message before sending. - Search and browse mail: Query by keywords, sender, receiver, time range, unread status, folder, tag, attachments, star, or importance, with automatic pagination. - Read full mail details: Retrieve bodies, attachments, inline images, and calendar/meeting invitations, including DLP encrypted-link scenarios. - Schedule and meeting emails: Send calendar invitations and meeting bookings through email with repeat rules and meeting options. - Use Case: Ask the agent to "reply to the Q2 report email from Zhang San thanking him and attaching the updated spreadsheet" — the Skill locates the mail, resolves recipients, writes the Markdown body, previews it, and sends the reply. ## Quick Start Ask the agent to search your WeCom mailbox for recent unread emails from a specific colleague and summarize their contents.

Frequently Asked Questions about wecomcli-email

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I send a WeCom email with attachments using wecom-cli?

Write the body to a local Markdown file, then call wecom-cli mail send with the file_path, subject, and recipients. Attachments accept either a media_id from prior upload or a local file_path that the CLI uploads automatically, with a 50MB total size limit.

How do I search WeCom emails by keyword, sender, or date?

Use wecom-cli mail search with JSON parameters such as keywords, sender, receiver, begin_time, end_time, only_unread, folder_names, or tag_names. Conditions combine with AND logic, and the --page-count flag fetches multiple result pages automatically.

Can this Skill mark emails as read, delete mail, or manage tags?

No. Marking read/unread, deleting emails, saving drafts, and writing tag operations are not supported by the interface. The Skill directs users to the WeCom client for those actions, though searching by folder or tag is supported.

How are inline images handled when sending WeCom emails?

Inline images use a placeholder written as ![]($name$) in the Markdown body, matched exactly by the content_id field in the inline_images array including the dollar signs. The image content is supplied via media_id or file_path, and any deviation breaks template matching.

What happens when an email contains DLP encrypted links?

When data-loss-prevention is enabled, attachments and inline images arrive as encrypted work.weixin.qq.com/filepreview/security URLs instead of media_id values. These links cannot be downloaded or decrypted by the CLI, so they are preserved as clickable links for the user to open in the WeCom client.

Does the Skill protect against malicious email content?

Yes. Email bodies are treated as data, never as instructions, so embedded prompt-injection text is ignored and flagged. The Skill also detects social-engineering patterns, warns about recipients extracted from email bodies, and refuses to write script tags or javascript URIs into outgoing mail.