Download File

Download large files from HTTP/HTTPS URLs with resume support, progress monitoring, and timeout handling

Security Scan

The skill is internally consistent with its stated purpose (downloading files with curl); it asks for nothing unexpected and contains only shell-based download instructions.

Purpose & Capability

Name/description match the instructions: the SKILL.md describes using curl (and optionally aria2) to download files, resume, show progress, and verify results. The only required binary is curl, which is appropriate and proportional.

Instruction Scope

Runtime instructions are narrowly focused on download-related actions (curl/aria2, polling sessions, ls/file/md5sum) and error handling. The guide does not instruct reading unrelated files, environment variables, or sending data to external endpoints other than the download URLs themselves.

Install Mechanism

No install spec or external downloads are defined; this is an instruction-only skill that expects curl to be available. That is the lowest-risk install profile.

Credentials

The skill requests no environment variables, credentials, or config paths. All required resources (curl, optional aria2) are reasonable and directly related to the stated functionality.

Persistence & Privilege

always is false and there are no elevated privilege requests. The skill assumes the agent can execute shell commands and poll background sessions (normal for an exec-capable agent). If you are concerned about autonomous downloads, consider limiting agent invocation scope or monitoring background processes.

Assessment

This skill appears to do exactly what it says — run curl (or aria2) to download files with resume and progress. Before installing: ensure you trust the agent and the URLs it will be asked to fetch (downloading arbitrary URLs can introduce malware), confirm curl/aria2 and any sandboxing/policy for exec are acceptable, and be aware large downloads consume disk and bandwidth. If you want tighter controls, ask for domain whitelisting, size limits, or require explicit user confirmation before each download.

📥 Large File Downloader

When to Use

Use this skill when downloading files from HTTP/HTTPS URLs, especially large files (>100MB).

Typical scenarios:

• Downloading software installers (DMG, EXE, ZIP)
• Downloading large datasets
• Downloading video/audio files
• Downloading documents/PDFs
• Any file download from URL

Trigger keywords:

• "download this file"
• "save this DMG"
• "download from this URL"
• "get this file"
• "帮我下载 xxx" (Chinese)

---

Core Principles

⚠️ Must-Follow Best Practices

1. Always use background execution - Never run large downloads synchronously
2. Always set timeout - Configure reasonable timeout based on file size
3. Always support resume - Use curl -C - for resumable downloads
4. Always show progress - Use --progress-bar flag
5. Always verify results - Check file size after download completes

---

Standard Process

Step 1: Start Download (Background Mode)

exec(
  command="curl -L -C - --progress-bar -o <destination_path> <URL>",
  background=true,
  timeout=600
)

Parameter explanation:

• -L - Follow redirects
• -C - - Resume partial download
• --progress-bar - Show progress bar
• -o - Output file path
• background=true - Critical! Run in background to avoid timeout
• timeout=600 - 10 minutes timeout (adjust based on file size)

Step 2: Monitor Progress

process(action="poll", sessionId="<session_id>")

Poll periodically to check download progress until completion.

Step 3: Verify Result

exec(command="ls -lh <file_path>")

Check:

• File exists
• File size is reasonable
• File is complete (compare MD5/SHA if available)

---

Complete Examples

Example 1: Download DMG File (215MB)

User: "Download https://github.com/example/app/releases/download/v1.0/app.dmg"

Agent action:

{
  "tool": "exec",
  "command": "curl -L -C - --progress-bar -o ~/Downloads/app.dmg https://github.com/example/app/releases/download/v1.0/app.dmg",
  "background": true,
  "timeout": 600
}

Returns: {status: "running", sessionId: "xxx"}

Then poll periodically:

{
  "tool": "process",
  "action": "poll",
  "sessionId": "xxx"
}

After download completes, verify:

{
  "tool": "exec",
  "command": "ls -lh ~/Downloads/app.dmg && file ~/Downloads/app.dmg"
}

Example 2: Download Small File (<50MB)

For small files, simplified handling is acceptable:

exec(command="curl -L -o ~/Downloads/small.pdf https://example.com/small.pdf", timeout=120)

Small files can run synchronously, but still set timeout.

---

Timeout Reference

File Size	Recommended Timeout
< 50MB	120 seconds (2 min)
50-200MB	300 seconds (5 min)
200-500MB	600 seconds (10 min)
500MB-1GB	1200 seconds (20 min)
> 1GB	1800 seconds (30 min) or more

---

Error Handling

Common Errors & Solutions

1. Download interrupted

# Re-run same command, curl -C - will resume automatically
curl -L -C - --progress-bar -o file.zip <URL>

2. Permission denied

# Ensure destination directory is writable
mkdir -p ~/Downloads

3. Redirect failed

# Use -L flag to follow redirects
curl -L -o file.zip <URL>

4. Network timeout

# Increase timeout or use --retry
curl -L --retry 3 --connect-timeout 30 -o file.zip <URL>

---

Advanced Options

Multi-threaded Download (aria2)

If aria2 is available, use multi-threading for faster downloads:

exec(
  command="aria2c -x 16 -s 16 -k 1M --continue -o ~/Downloads/file.zip <URL>",
  background=true,
  timeout=600
)

Full Flow with Progress Monitoring

# 1. Start download
exec(command="curl -L -C - --progress-bar -o ~/Downloads/file.zip <URL>", background=true, timeout=600)

# 2. Poll every 30 seconds
process(action="poll", sessionId="xxx")

# 3. Verify after completion
exec(command="ls -lh ~/Downloads/file.zip")

# 4. Optional: Calculate checksum
exec(command="md5sum ~/Downloads/file.zip")

---

Security Considerations

1. Download from trusted sources only - Verify URL origin
2. Check file type - Use file command to verify
3. Scan for viruses - If needed, scan after download
4. Don't overwrite important files - Check if destination exists

---

Related Skills

• feishu-send-file - Send downloaded file to Feishu
• file-read - Read downloaded file content
• video-frames - Extract frames if video file

---

Troubleshooting

If download fails, follow these steps:

1. Check network connectivity

   curl -I <URL>
   

2. Check disk space

   df -h ~/Downloads
   

3. View detailed error

   curl -v -o /dev/null <URL>
   

4. Try alternative tool

   wget --continue <URL>
   

---

Remember: Always use background=true for large file downloads! 🎯