Token导航 LogoToken导航TokenDH.com
研究检索需要联网clawhub未标认证来源可访问clear审计提醒

auto-file-sender自动文件发送器

Agent Skill

auto-file-sender 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

6,120

周安装

250

GitHub Stars

公开资料未说明

下载量

1,960
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:auto-file-sender(自动文件发送器)
来源仓库:https://github.com/timyljob2011-sudo/auto-file-sender
安装命令:
openclaw skills install auto-file-sender
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install auto-file-sender

简介

auto-file-sender 用于查找、检索和筛选相关信息,适合在 OpenClaw 中根据关键词快速定位候选结果时使用。

  • 支持自动发送文件到 Feishu/Lark,适用于文档生成后即时分发的场景。
  • 通过 clawhub 安装,命令为 openclaw skills install auto-file-sender,需结合来源仓库进一步确认具体用法。
  • 安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写操作。
  • 当前功能描述基于原始 README,实际能力以官方文档为准。

SKILL.md

name
auto-file-sender
description
|

Auto File Sender

Overview

This skill enables automatic file delivery from the workspace to Feishu/Lark users. When files are generated (documents, PDFs, images, etc.), they can be automatically sent to specified recipients without manual intervention.

Key Capabilities:

  • Auto-detect new files in workspace
  • Send via Feishu message with file attachment
  • Support batch sending of multiple files
  • Configurable file type filters and recipient rules

Quick Start

Basic Usage

When a file is ready to send:

// Single file
{
  "action": "send",
  "filePath": "/root/.openclaw/workspace/document.docx",
  "filename": "document.docx",
  "message": "Here's your file!",
  "target": "user_open_id"
}

Auto-Send on File Creation

The skill provides a helper script to watch for new files and auto-send:

# Watch workspace and auto-send new files
python3 scripts/auto_send.py --watch /root/.openclaw/workspace --recipient USER_OPEN_ID

Workflow

Step 1: Identify Files to Send

Check for recently created/modified files:

# List files created in last 10 minutes
find /root/.openclaw/workspace -type f -mmin -10

Step 2: Send Files

Use the message tool with filePath parameter:

{
  "action": "send",
  "filePath": "<absolute-path-to-file>",
  "filename": "<display-filename>",
  "message": "<optional-message>",
  "target": "<recipient-open-id>"
}

Parameters:

  • filePath: Absolute path to the file (required)
  • filename: Display name for the file (optional, defaults to basename)
  • message: Accompanying text message (optional)
  • target: Recipient open_id (defaults to current user if omitted)

Step 3: Confirm Delivery

Check the response for successful delivery:

  • messageId: ID of the sent message
  • chatId: ID of the chat/channel

Supported File Types

TypeExtensionsMax Size
Documents.docx, .doc, .pdf30MB
Images.jpg, .png, .gif, .webp30MB
Spreadsheets.xlsx, .xls, .csv30MB
Archives.zip, .tar.gz30MB
OthersAny30MB

Batch Sending

To send multiple files at once:

// Send files sequentially
for (const file of files) {
  await message.send({
    action: "send",
    filePath: file.path,
    filename: file.name
  });
}

Configuration

Default Settings

  • Source directory: /root/.openclaw/workspace
  • Max file size: 30MB (Feishu limit)
  • Auto-recipient: Current conversation user

Custom Recipient

To send to a specific user:

{
  "action": "send",
  "target": "ou_a65105519c863f8544fb22b40c468063",  // User's open_id
  "filePath": "/path/to/file"
}

Scripts

scripts/auto_send.py

Python script for watching directories and auto-sending files.

Usage:

python3 scripts/auto_send.py [options]

Options:
  --watch PATH       Directory to watch (default: workspace)
  --recipient ID     Target recipient open_id
  --pattern PATTERN  File pattern to match (default: *)
  --once             Send existing files and exit (don't watch)

Examples:

# Watch and auto-send all new PDFs
python3 scripts/auto_send.py --pattern "*.pdf" --recipient USER_ID

# One-time send of all docx files
python3 scripts/auto_send.py --pattern "*.docx" --once

Troubleshooting

File Not Found

  • Ensure file path is absolute
  • Verify file exists: ls -la <filepath>
  • Check file permissions

Send Failed

  • Verify file size < 30MB
  • Check recipient open_id is correct
  • Ensure bot has permission to send files

Large Files

For files > 30MB:

  1. Compress: zip -r output.zip large_file
  2. Split: split -b 25M large_file part_
  3. Use cloud storage and send link instead

Best Practices

  1. Always verify files exist before sending
  2. Use descriptive filenames for better organization
  3. Batch similar files to reduce API calls
  4. Clean up sent files periodically to save space
  5. Log sent files for tracking (optional)

Examples

Example 1: Send Generated Document

User: "Generate a report and send it to me"

// After generating the report
{
  "action": "send",
  "filePath": "/root/.openclaw/workspace/report_2024.docx",
  "filename": "Annual_Report_2024.docx",
  "message": "Here's your annual report!"
}

Example 2: Send Multiple Files

User: "Send all the PDFs in my workspace"

# Find and send all PDFs
find /root/.openclaw/workspace -name "*.pdf" -exec \
  python3 -c "import sys; print(sys.argv[1])" {} \;

Then send each file using the message tool.

Example 3: Auto-Send on Completion

After a long-running task generates output:

// Task completed, auto-send result
{
  "action": "send",
  "filePath": "/root/.openclaw/workspace/output.pdf",
  "message": "Task completed! Here's your file."
}

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

85.05%
按下载量换算1,667

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills