AI-Powered File Transfer Automation for Post-Production

Post-production generates enormous volumes of files that need to move between people, departments, and systems. Traditionally, this meant:

• Manual transfers: Someone copies files and sends links
• Watch folders: Cloud transfer services monitor folders and trigger uploads
• Scripts: Custom automation that breaks when anything changes

All of these have the same limitation: they're dumb. Watch folders don't understand what the files are. Scripts can't adapt to new situations. Manual processes don't scale.

Enter AI Agents

Handrive exposes a 43-tool MCP server that AI agents like Claude Code can use to fully automate file workflows. Unlike watch folders, AI agents understand context:

• Read file names and metadata to understand what files are
• Make decisions about routing, naming, and permissions
• Handle exceptions and edge cases intelligently
• Coordinate complex multi-step workflows

What's Possible with 43 MCP Tools

Handrive's MCP server provides tools across 8 categories:

Example: Automated Dailies Ingest

Here's how an AI agent can automate the entire dailies ingest workflow:

# Claude Code prompt:
"Monitor /Volumes/DIT_Drive for new camera card folders.
When a new folder appears:
1. Parse the folder name to extract shoot date and camera
2. Create a share named 'Dailies - [Date] - [Camera]'
3. Add the editor (editor@studio.com) with Editor role
4. Add the director (director@studio.com) with Viewer role
5. Rename files using our naming convention: [Project]_[Date]_[Camera]_[Clip]
6. Organize into scene folders based on slate/metadata
7. Send a Slack notification when complete"

# The AI agent uses Handrive MCP tools to execute this:
- list_files: Monitor the DIT drive
- create_share: Create new share with parsed name
- add_member: Add editor and director
- create_file/rename_file: Organize and rename
- External: Send Slack notification

Compare to Watch Folders

Example: Multi-Department Delivery Pipeline

# Claude Code prompt:
"When the editor marks a sequence as 'locked':
1. Find the locked sequence in the Exports share
2. Create a new share 'VFX Handoff - [Sequence Name]'
3. Add the VFX supervisor with Editor role
4. Export a list of VFX shots from the project metadata
5. Create placeholder folders for each VFX shot
6. Send email to VFX team with share link and shot list

When VFX returns final shots:
1. Detect new files in the VFX returns folder
2. Verify frame counts match expected
3. Add to the Color share for grading
4. Remove VFX supervisor access (they're done)
5. Add colorist with Editor role
6. Send notification to colorist"

This kind of context-aware, multi-step workflow is impossible with traditional watch folders. The AI agent understands the production pipeline and orchestrates accordingly.

Setting Up AI Automation

Prerequisites

• Handrive installed and authenticated
• Handrive server running (desktop or headless)
• Claude Desktop or Claude Code with MCP enabled

Configuration

Add Handrive to your Claude Desktop config:

{
  "mcpServers": {
    "handrive": {
      "command": "/path/to/handrive",
      "args": ["mcp"]
    }
  }
}

Test the Connection

Ask Claude: "List all my Handrive shares"

If connected, Claude will use the list_shares tool and show your shares. From there, you can build complex automations.

Real-World Automation Ideas

• Automated backup: AI monitors project folders, creates backup shares, manages version history
• Client delivery portal: AI creates shares per client, sets expiration dates, sends branded notifications
• Review workflow: AI routes cuts to reviewers based on project stage, collects approvals, moves to next stage
• Archive management: AI identifies completed projects, moves to archive shares, manages storage lifecycle
• Multi-site sync: AI coordinates transfers between multiple Handrive servers at different locations

Why This Matters

Post-production is increasingly AI-augmented — AI upscaling, AI-driven VFX, AI-assisted editing. The file management layer needs to keep up. Traditional tools can't handle the volume, complexity, or speed required.

Handrive's MCP server makes file operations first-class citizens in AI workflows. The same AI that helps you edit can also manage your entire file pipeline.