← 返回 Skills 市场

Ali Feishu Sender

Name: Ali Feishu Sender
Author: yuan-huicheng

作者 yuan-huicheng · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

100

总下载

当前安装

版本数

在 OpenClaw 中安装

/install ali-feishu-sender

功能描述

Send multimedia content to Feishu (Lark) via API. Use when: (1) sending images, audio, video, rich text, or cards to Feishu users/groups, (2) uploading media...

使用说明 (SKILL.md)

Feishu Media Sender

Send multimedia messages to Feishu with proper formatting: inline images, voice bubbles, video players, rich text, and cards.

Dependencies

ffmpeg at D:\ffmpeg\bin\ffmpeg.exe (for audio conversion, video fix)
ffprobe at D:\ffmpeg\bin\ffprobe.exe (for duration detection)
Python 3.7+
Feishu app credentials (pre-configured in script defaults)

Quick Reference

Run the script at scripts/feishu_media.py:

import sys; sys.path.insert(0, r'\x3Cskill_dir>/scripts'); from feishu_media import *

Available Functions

Function	Message Type	Key Features
`send_text(text)`	text	Plain text
`send_image(filepath)`	image	Inline embed, auto-detect mime
`send_audio(filepath)`	audio	Voice bubble, auto→opus, auto duration
`send_video(filepath, cover_image=None)`	media	Inline player, faststart, auto duration
`send_rich_text(title, elements)`	post	Rich text with embedded media
`send_card(title, elements)`	interactive	Card with header color

All functions accept optional: open_id, token, app_id, app_secret.

Key Rules (learned from testing)

Audio: Must be opus format, must include duration (ms). Auto-converts non-opus via ffmpeg.
Video: Must upload as file_type=mp4, must include duration (ms). Script auto-applies -movflags +faststart.
Video message type: Use msg_type=media (not video), content uses file_key + optional image_key for cover.
Image in video: image_key is the cover thumbnail. Without it, video has no preview frame.
Rich text media: Use {'tag': 'media', 'file_key': '...', 'image_key': '...'} as an independent paragraph.
File upload types: opus (audio), mp4 (video), stream (generic file), pdf, doc, xls, ppt.

Example: Send MiniMax content to Feishu

# Image
send_image('minimax-output/kitten.png')

# TTS voice bubble (auto converts mp3→opus)
send_audio('minimax-output/greeting.mp3')

# Video with cover (auto faststart + duration)
send_video('minimax-output/kitten_video.mp4', cover_image='minimax-output/kitten.png')

# Rich text with video
send_rich_text('阿离的报告', [
    [{'tag': 'text', 'text': '一舞剑气动四方~'}],
    [{'tag': 'media', 'file_key': 'file_xxx', 'image_key': 'img_xxx'}],
])

CLI Usage

python scripts/feishu_media.py text "Hello"
python scripts/feishu_media.py image photo.png
python scripts/feishu_media.py audio greeting.mp3
python scripts/feishu_media.py video clip.mp4 --cover thumb.png

安全使用建议

Before installing or using: 1) Don’t hand over Feishu app_id/app_secret unless you trust the skill — the script will use these to request a tenant token and then upload files to Feishu. Prefer supplying a short-lived token if possible. 2) Verify and, if needed, correct the ffmpeg/ffprobe paths — the code defaults to D:\ffmpeg\bin\ which is Windows-specific and may not exist; ffmpeg is required for audio/video conversion. 3) Note the registry metadata omits the env vars and binary dependencies declared in SKILL.md/code — ask the publisher to correct the manifest so required credentials and binaries are explicit. 4) Review the script in a safe environment (e.g., sandbox or isolated VM) with non-sensitive test files to confirm behavior. 5) Be aware the tool will read local media files and upload them to Feishu — do not use with private data unless you intend to store/send it to your Feishu tenant. 6) If you need higher assurance, request the publisher to (a) document exact env vars, (b) remove hard-coded paths or make them configurable, and (c) sign or publish the package from a known source so you can track updates.

功能分析

Type: OpenClaw Skill Name: ali-feishu-sender Version: 1.0.0 The skill is a legitimate utility for sending multimedia content to the Feishu (Lark) platform via its official API (open.feishu.cn). It uses subprocess calls to ffmpeg and ffprobe for media processing (audio conversion and video optimization) and handles authentication via standard environment variables (FEISHU_APP_ID, FEISHU_APP_SECRET), with no evidence of malicious intent, unauthorized data exfiltration, or obfuscation in scripts/feishu_media.py or the SKILL.md instructions.

能力评估

ℹ Purpose & Capability

The code implements Feishu media upload and messaging endpoints (images, audio→opus conversion, mp4 faststart, rich text, cards) which matches the skill description — functionality is coherent with the stated purpose. However the SKILL registry lists no required env vars/binaries while SKILL.md and the script expect Feishu credentials and ffmpeg/ffprobe (Windows path D:\ffmpeg\bin), which is an inconsistency.

✓ Instruction Scope

Runtime instructions and the script focus on reading local media files, converting/inspecting them via ffmpeg/ffprobe, obtaining a tenant access token from Feishu, uploading files/images, and sending messages to Feishu APIs. There are no instructions to read unrelated system files, contact unexpected external endpoints, or exfiltrate data outside the Feishu APIs.

✓ Install Mechanism

This is an instruction-only skill with a single Python script and no install spec, so nothing is downloaded or installed by the skill itself. That lowers install risk. The script expects external binaries (ffmpeg/ffprobe) to exist on the host.

⚠ Credentials

Registry metadata claims no required environment variables, but the script reads FEISHU_APP_ID, FEISHU_APP_SECRET, and FEISHU_OPEN_ID from the environment (and will call the token endpoint if no token is provided). The SKILL.md also expects ffmpeg/ffprobe at a specific Windows path. Requesting Feishu app credentials is reasonable for this purpose, but the registry should declare them. The mismatch is a practical and security concern (users may unknowingly provide credentials).

✓ Persistence & Privilege

The skill does not request persistent/always-on privileges (always:false) and does not modify other skills or system-wide settings. It performs network calls only to documented Feishu API endpoints.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install ali-feishu-sender
安装完成后，直接呼叫该 Skill 的名称或使用 /ali-feishu-sender 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release of ali-feishu-sender. - Send multimedia (images, audio, video, rich text, interactive cards) to Feishu (Lark) users or groups via API. - Supports inline images, native voice bubbles (opus+duration), inline video players (mp4+duration+cover), and rich text with embedded media. - Automatically handles media format requirements (e.g., audio to opus, video faststart, duration extraction). - Provides functions and CLI to send content and upload media to Feishu. - Designed for posting generated content (e.g., MiniMax images, audio, video) to Feishu easily.

元数据

Slug ali-feishu-sender

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题