CONTENT CREATION

Media librarian agent: caption, dedupe, and organize Dropbox assets into Airtable

An agent that reviews a Dropbox media backlog, captions and zero-shot tags each image with Hugging Face, detects likely duplicates, and curates a clean.

CategoryContent Creation
Enginepaperclip
Difficultyadvanced
Triggerschedule
Steps6
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerScheduled or manual agent run
  • ActionList and read new images from DropboxDropboxDropbox
  • ActionCaption and zero-shot tag via Hugging FaceHugging FaceHugging Face
  • LogicReason over duplicates and target collection
  • ActionUpsert curated records into AirtableAirtableAirtable
  • OutputPost run summary report to SlackSlack

What it does

This runs an agent that acts like a media librarian over a Dropbox backlog. For each image it generates a caption and zero-shot category tags via Hugging Face, reasons about near-duplicates and naming, decides the right Airtable collection, and files a curated record. At the end it writes a summary of what it captioned, tagged, flagged as duplicate, and what needs human attention.

When to use it

A messy shared Dropbox of mixed assets needs to become an organized, captioned, deduped Airtable catalog — and the cleanup involves judgment (is this a dup, which collection, is the caption good enough) rather than a fixed rule set.

How it works

  1. 1A schedule or manual run starts the agent on the Dropbox folder.
  2. 2The agent lists and reads new images from Dropbox.
  3. 3For each image it calls Hugging Face for a caption and zero-shot tags.
  4. 4It reasons over captions and tags to detect probable duplicates and pick a target collection.
  5. 5It upserts curated records into Airtable, flagging duplicates and weak captions.
  6. 6It posts a run summary report to Slack for the content owner.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect DropboxFiles and folders.
  2. 2
    Connect Hugging FaceModels, datasets, spaces — the open-source hub.
  3. 3
    Connect AirtableBases, tables, views, automations.
  4. 4
    Connect SlackChannels, DMs, threads, mentions.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.