feature: Cloud-based file organizer agent

[EthanL3]

Feature Description

Cloud storage buckets (e.g., GCS, S3) quickly become cluttered as users and teams upload files without consistent structure. Manually sorting or tagging files is time-consuming, error-prone, and unscalable.

Proposed Solution

Implement a cloud-based File Organizer Agent that automatically classifies and routes newly uploaded files based on configurable rules and semantic inference.

Workflow
The agent will do the following:

1. Monitor bucket upload events (e.g., Pub/Sub or S3 notifications)
2. Match against user-defined rules by checking regex against the uploaded filenames
3. If no rule applies, then semantically classify using pre-trained model
4. Move or copy files to appropriate path

Specifications:
Example user-defined rule config:

rules:
  - pattern: "*.csv"
    target: "~/Projects/Data"
    action: "move"
    priority: 10

  - pattern: ["*.py,", "*.java"]
    target: "~/Code"
    action: "move"
    priority: 20

• Higher priority takes higher precedence if both/multiple rules apply
• Always tries to use these rules before classification
• Will have "dry run" mode where user can what actions agent would take before enabling it fully
• Optionally: can tag certain files instead of in addition to moving them (e.g. "Team1", "Product", "Business Docs")

APIs:

• SentenceTransformers: Used for filename classification, identifies category
  • Optionally: scan file text content to further infer category
• Cloud Storage APIs: Watch for object creation events, moving/copying files

Want to contribute?

• I would like to work on this issue.

Additional Context

No response