A specialized tool for AI agents to programmatically send images and files to Feishu (Lark) users and group chats.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install feishu-send-image
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 feishu-send-image 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 Feishu Send Message?
The Feishu Send Message skill is a robust integration designed for the Feishu (Lark) platform, allowing AI agents to communicate beyond simple text. By leveraging Openclaw Skills, developers can enable their agents to share visual assets and documents directly within the Feishu workspace. It abstracts the complexity of the Feishu Open API, handling everything from authentication and token management to media uploading and message formatting.
This skill is particularly valuable for teams that rely on Feishu for internal operations, as it allows for the creation of automated reporting bots, asset delivery pipelines, and collaborative workflows. Whether you are building an agent to share data visualizations or distribute project files, this skill provides the necessary bridge between your local environment and the Feishu ecosystem.
Feishu Send Message Use Cases
- Automatically sending generated charts or UI mockups to a specific developer open_id.
- Distributing daily PDF reports or build artifacts to a team group chat via chat_id.
- Responding to agent prompts like "share this image" or "send the document to the team" during an interactive session.
- Integrating automated notifications with media attachments into a CI/CD pipeline managed by Openclaw Skills.
How Feishu Send Message Works
- The skill reads the appId and appSecret from the local configuration file located at ~/.openclaw/openclaw.json.
- It communicates with the Feishu Authentication API to retrieve a valid tenant_access_token.
- Based on the specified message type, it uploads the target file or image to Feishu's servers to obtain a unique image_key or file_key.
- If an image is being sent on macOS, it utilizes the sips utility to ensure the image meets Feishu's resolution and size constraints.
- The skill sends the final message payload containing the media key to the specified recipient ID (either a user or a group chat).
Feishu Send Message Setup
To get started with this skill, ensure you have a Feishu Custom App with the necessary permissions. Store your credentials in your local config file:
# Edit ~/.openclaw/openclaw.json
{
"appId": "cli_xxxxxxxx",
"appSecret": "xxxxxxxxxxxxxxxx"
}
Ensure that curl, stat, and sips (for macOS users) are available in your environment. You can then trigger the skill via the CLI:
feishu_send_message --msg-type image --file-path ./chart.png --receive-id ou_12345 --receive-id-type open_id
Feishu Send Message Data Schema & Taxonomy
The skill utilizes a structured command-line interface and requires specific metadata to function within Openclaw Skills.
| Parameter | Description | Required | Options |
|---|---|---|---|
| --msg-type | The category of message to send | Yes | image, file |
| --file-path | The local filesystem path to the asset | Yes | String (path) |
| --receive-id | The unique identifier for the recipient | Yes | open_id, chat_id |
| --receive-id-type | The type of recipient ID provided | No | open_id (default), chat_id |
File Constraints:
- Images: Maximum 10MB, Resolution < 12000x12000.
- Files: Maximum 100MB.
Feishu Send Message Advanced Features
- Automatic image optimization: Uses native macOS sips tools to compress and resize images if they exceed platform limits.
- Intelligent file type detection: Automatically determines file_type (e.g., pdf, doc, mp4) based on file extensions to ensure correct rendering in the Feishu client.
- Multi-recipient support: Flexible ID typing allows for seamless switching between direct user messaging and group channel broadcasting.
- Stream-based uploads: Handles unknown file formats as generic streams to ensure maximum compatibility for all document types within Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
OpenClaw Message Delivery Guide
A technical framework for AI agents to ensure asynchronous tasks, background processes, and reminders actually reach the user through verified delivery paths.
aliyun-clawscan
A specialized security assessment tool designed to audit OpenClaw configurations and perform deep static analysis on skills to detect vulnerabilities and malicious behavior.
Skill Vetting
A specialized security framework designed to audit and validate third-party AI skills for safety, utility, and prompt injection risks.
WhisperX Speech Recognition
A high-performance local speech recognition skill that provides lightning-fast offline transcription with word-level precision.
GDrive: Approved-Folder Google Drive Access
A privacy-first Google Drive integration that restricts agent access to a pre-defined allow-list of approved folders.
MakeAIClips — AI Video Clipper
MakeAIClips is a powerful AI-driven tool that automatically converts long-form YouTube videos into high-engagement vertical clips with burned-in captions and viral hook titles.
Featured*
