Coming soon · Bucket Intelligence

    Turn any bucket into a queryable knowledge base

    Auto-index your files as they arrive. Ask questions in plain language — powered by your own OpenAI, Anthropic, or Cohere keys.

    1-click

    Enable on any bucket

    No migration or data export. Works with your existing storage plan.

    $0

    LLM markup

    Bring your own OpenAI, Anthropic, or Cohere keys. Provider costs go directly to you.

    < 1 min

    Indexing time

    New uploads are chunked and indexed in near real-time as they arrive.

    Features

    Built around your existing stack

    No chunking scripts, no vector DB to provision, no new API keys to manage — just your bucket and your LLM provider.

    Bring your own LLM keys

    Use OpenAI, Anthropic, or Cohere for embeddings and completions. You pay your provider directly — no markup from Fil One.

    Automatic indexing

    Files uploaded to enabled buckets are indexed in near real-time. Deleted files are automatically removed from the index.

    Multiple integration options

    Access your knowledge base via the Fil One dashboard, MCP server for Claude and Cursor, or direct REST API.

    Custom embedding models

    Use any OpenAI-compatible endpoint, including self-hosted models. Switch providers without re-indexing.

    Supported file types

    PDFMarkdownPlain textHTMLDOCX
    How it works

    Up and running in minutes

    01

    Connect your bucket

    Enable Bucket Intelligence on any existing bucket with one click. No migration or data export needed.

    02

    Files get indexed automatically

    New uploads are chunked and indexed in near real-time using your chosen embedding model and API key.

    03

    Ask questions, get answers

    Query via the dashboard, MCP endpoint, or REST API. Answers come with cited source file excerpts.

    FAQ

    Common questions

    PDF, Markdown, plain text (.txt), HTML, and DOCX at launch. More formats on the roadmap.

    Encrypted at rest using industry-standard encryption. Keys are never logged or included in responses.

    Near real-time — typically under a minute for most files. Larger documents may take a few minutes.

    Yes. Any OpenAI-compatible endpoint works, including self-hosted models like nomic-embed or mxbai-embed.

    It is automatically removed from the index. No manual cleanup needed.

    Turn your buckets into knowledge bases

    Early access is open. Join the waitlist and be first in line.

    Requires an active storage plan · No extra infrastructure